Slashdot Mirror


AMD Launches Ryzen, Claims To Beat Intel's Core i7 Offering At Half the Price (hothardware.com)

Reader MojoKid writes: AMD CEO, Dr. Lisu Su took to the stage at AMD's Ryzen tech day yesterday and opened the event with official speeds, feeds, pricing, and benchmark scores for the company's upcoming Ryzen series processors. AMD's goal with Ryzen, which is based on its Zen microarchitecture, was a 40% IPC (instructions per clock) uplift. As it turns out, AMD was actually able to increase IPC by approximately 52% with the final shipping product, sometimes more depending on workload type. Dr. Su also showed the first die shot of an 8-core Ryzen processor, disclosing that it consists of approximately 4.8 billion transistors. AMD's flagship Ryzen 7 1800X 8-core/16 thread CPU will have a base clock speed of 3.6GHz, a boost clock of 4.0GHz, and a 95 watt TDP. AMD claims the Ryzen 7 1800X will be the fastest 8-core desktop processor on the market when it arrives. The next member of the line-up is the Ryzen 7 1700X with a base clock of 3.4GHz and a boost clock of 3.8GHz, also with 8 cores and a 95 watt TDP. Finally, the Ryzen 7 1700 – sans X – is also an 8-core / 16-thread CPU, but it has lower 3.0GHz base and 3.7GHz boost clocks, along with a lower 65 watt TDP. AMD took the opportunity to demo the Ryzen 7 1800X and it was approximately 9% faster than the Core i7-6900K running Cinebench R15's multi-threaded test, at about half the cost. And in another comparison, Dr. Su put the 8-core 7 1700 up against the quad-core Core i7-7700K, converting a 4K 60 FPS video down to 1080P and the Ryzen CPU outpaces the Core i7 by 10 full seconds. Pricing for the three initial Ryzen 7 series processors will undercut competing Intel processors significantly. AMD's Ryzen 7 1800X will arrive at $499, Ryzen 7 1700X at $399, and Ryzen 7 1700 at $329. Based on current street prices, Ryzen will be between 20% — 50% lower priced but AMD is claiming performance that's better than Intel at those price points.

281 comments

  1. FINALLY! by Anonymous Coward · · Score: 5, Insightful

    Finally competition from AMD! Stop this stagnation madness!

    1. Re:FINALLY! by Cederic · · Score: 1, Interesting

      Yeah, I fell out of love with AMD due to their inability to compete on power/performance grounds but I'd very much welcome a serious competitor to Intel.

      Similar to the AMD graphics cards. I'm unlikely to buy anything called 'Radeon' but I'm glad they exist and force Nvidia to continue to improve and innovate.

      If these new AMD chips provide comparable performance at comparable heat levels (I don't care about the power used, I care about the noise needed to dissipate the heat generated) then I'm more than happy to switch back to AMD and save some cash.

    2. Re:FINALLY! by aliquis · · Score: 2

      It's not ALL that much though.

      Like the 1700X vs 6800K both at $400.

      One finish in 100 seconds the other in 112 seconds.
      That's 11% faster, but with 33% more cores and a lower / core performance.

      The major advantage there though is that a B350 motherboard you will likely be able to get for $100 whereas a X99 one will cost $200.

      1800X vs 6900K is half-price for similar performance so that's of course massive but they aren't all that much faster than the cheaper processors. AMD just doesn't charge as much premium for it.

      Ryzen 7 1700 vs i7 7700K at about the same price.
      8 cores at 3 GHz vs 4 cores at 4 GHz.
      You get 46% higher multi-core performance, expected at same IPC and clock would had been ~50% so very close performance per clock and core but then there's the fact of the clock speed difference and hence the 7700K most likely being faster in single-core tasks. So the question then becomes if you want 25-26% lower single-core performance or 46% higher multi-core performance.

      It's good competitive products and very nice processors for multi-threaded tasks and needs but there's still room for Intel to be relevant at-least.

    3. Re: FINALLY! by Bing+Tsher+E · · Score: 0

      Competition is good.

      Can AMD compete, though? One would hope so, but they will need more than a product. They will need motherboard vendors to provide sockets, and developers to write code that takes advantage of the edge their processor gives.

      Fanboys can tag along, like the dudes who put racing additive stickers on their rusty Hondas.

    4. Re:FINALLY! by Big+Hairy+Ian · · Score: 4, Insightful

      Finally competition from AMD! Stop this stagnation madness!

      Don't worry the MBA's will find some way to screw this up they always do

      --

      Build a Man a Fire, and He'll Be Warm for a Day. Set a Man on Fire, and He'll Be Warm for the Rest of His Life.

    5. Re:FINALLY! by Anonymous Coward · · Score: 0, Insightful

      AMD has been solid for the past decade or more. Your religion is only costing you money.

    6. Re:FINALLY! by MachineShedFred · · Score: 1, Interesting

      Yeah, I love spending my money on inferior products that don't work as well, because brand loyalty!

      It's called 'voting with your wallet.' If the company isn't getting it done with their product line, they need to know it. And there's no better message delivery than a warehouse full of stale product that nobody wants to buy. How do we know? Because of this story right here - AMD may have come up with a winner. Same or better performance at far less price. And why did they do this? Because Intel was kicking their asses up and down the block on performance, at ANY price point.

      Maybe not anymore. If AMD can deliver on this announcement, they will see sales uplift. Message received.

      --
      Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    7. Re: FINALLY! by snookiex · · Score: 1

      I'm really rooting for AMD. Some of the first computers I used had K5/K6 chips, they were good, but still outperformed by the equivalent Pentiums. Later, many AMD processors for laptops suffered overheating problems (and all the laptops I wanted had Intel processors). Even though, as you say, competition is good and I'm really looking forward to getting an AMD-powered pc in the near future if there's a good enough one around.

      --
      Open Source Network Inventory for the masses! Kuwaiba
    8. Re:FINALLY! by Anonymous Coward · · Score: 0

      > Yeah, I fell out of love with AMD due to their inability to compete on power/performance grounds [...]

      So you are one of those consumerist sheep following the easiest and cheapest. No spine, no own convictions. "Fell out of love" my ass, simply dropping support because other alternatives were easier.

      The ones like you are fully responsible for the dire situation we have these days. Thankyouverymuch.

      Grow a spine!

      I, for one, buy what is best for myself, not for the company that makes the item. I have no loyalty to any particular company. Business is business. If a company delivers, I'll buy their products. If another company delivers better, I'll buy their products instead. That's what a free market economy is all about. Loyalty? Don't make me laugh.

    9. Re: FINALLY! by dougdonovan · · Score: 1

      keep an eye on their stocks

    10. Re:FINALLY! by Anonymous Coward · · Score: 0

      "Yeah, I love spending my money on inferior products that don't work as well, because brand loyalty!"

      Don't know what bullshit you're on about. I'm still playing just about any game I want on an Athlon II X4 640. That includes newer releases such as The Culling.

    11. Re: FINALLY! by Bing+Tsher+E · · Score: 1

      I think the first motherboard I ever owned had an AMD 8088 processor chip on it. TURBO! It was 8 Mhz, not one of those slow Intel 4.77 Mhz jobbies.

    12. Re:FINALLY! by ezelkow1 · · Score: 1

      Wow

      You call him a consumerist sheep, yet you are the one supporting the, i guess from your comment, harder and more expensive solution. For what gains? do you own stock in AMD? Do you work at AMD? No? then why would you do that yourself.

    13. Re:FINALLY! by Anonymous Coward · · Score: 0

      If x265 ever takes off, you'll need something
      like this to play 4k videos.

      CAP === 'strides'

    14. Re:FINALLY! by Anonymous Coward · · Score: 1

      Beware the hype train - And this is a hype train of the strongest degree. We're approaching No Man's Sky levels of hype here.

      Keep in mind this is a soft launch. No launch silicon in reviewers hands.No objective reviews. Only "benchmarks" that came from AMD, which will always paint their product in a good light.

      Intel isn't "stagnating" - They're responding to market forces. You really have not needed a faster CPU since the launch of Sandy bridge. What the market wants is more features, lower power. That is what Intel has been focusing on and delivering pretty well.

      Lastly, remember that single thread performance is still king in the desktop space. Lots of cores are great for servers and some applications - But for user facing applications (Including games) the most heavily weighted cpu performance bottleneck is the top speed of your core(s). - Multi-threaded programming is hard and there is no magical compiler switch or library that will suddenly make lots of slow cores = One fast core. (For most applications)

      We all want AMD to give Intel some competition and to push prices down.. But don't hold your breath until reviewers have shipping parts in their hands.

    15. Re: FINALLY! by haruchai · · Score: 2

      I'm really rooting for AMD

      Not only am I rooting for them, most of my PCs since the late 90s have been AMD whereas all my laptops and the majority of my servers have been Intel (with a few Sparcs & IBM Power thrown in). As AMD was lagging the past few years, I was seriously thinking of buying Intel for my next performance workstation.
      Now it looks like I can continue with AMD a while longer and if Ryzen measures up, it'll be my 2017 Xmas present to myself.

      --
      Pain is merely failure leaving the body
    16. Re: FINALLY! by Anonymous Coward · · Score: 0

      I had a K6 and an intel pentium chip at the same time similar pricepoint and the amd at the time was far and above the better chip. this was around 95 96 if i recall correctly

    17. Re:FINALLY! by Anonymous Coward · · Score: 1

      I vote against Intel with my wallet. It's nice that I can get a better, cheaper product for that now, too.

    18. Re: FINALLY! by Anonymous Coward · · Score: 0

      Generally agree with you but power=heat in this case.

    19. Re:FINALLY! by Anonymous Coward · · Score: 0

      All my systems were AMD but i was tired of waiting to find out IF or WHEN they would release a new CPU and i did not want to wait to find out it was trash, so for the first time i went with Intel last November i still think i made the right choice. It always seemed to me Intel boards had the options i wanted and AMD were very limited.

    20. Re: FINALLY! by CrashNBrn · · Score: 3, Interesting
      The only "Intel inside" that were allowed in our house are in the wife's Macs... I'll buy Intel when they are the only option left.

      AMD 8088, AMD 386, AMD K6-2, Athlon S754, Phenom II X2 560 Black Edition, FX-6300 Black Edition, an AMD A8 and A10-8700P laptop.

      All but the first three are still in working order. Looking forward to building a new Ryzen desktop later this year.

    21. Re:FINALLY! by Anonymous Coward · · Score: 0

      Grow a spine!

      Says the AC!

    22. Re:FINALLY! by nobuddy · · Score: 5, Interesting

      AMD- lately- is 90% of the performance at half (and often less) the price. I'll take that any day. My 8300 with R9 270x plays every game in existence without a problem. Well, it stutters a little on Star Citizen. But so does every PC ever made, no matter how much of a gaming beast it may be.

    23. Re:FINALLY! by Anonymous Coward · · Score: 0

      The problem with AMD hardware is that their CPUs don't support virtualization properly. I have 2 FX-8350s and they can't run the Android Development Kit emulators; some instruction is not supported. I had another problem with them a few years ago on a contract trying to set up some SQL server replication; again some not supported instruction issue. I give up - Intel it will be until these guys get their act together. Maybe these new chips will support all types of virtualization, but that better be advertised front row center in all of their communications with provable benchmarks and reviews before I even consider them again for anything. And that goes for their GPUs as well; NVidia smokes them currently.

    24. Re:FINALLY! by ottdmk · · Score: 1

      I have to agree. I went with an AMD FX-6300 for my last computer. Definitely saved money, and I have no qualms about the performance. Does a fantastic job (coupled with 8gig of RAM) at compiling LibreOffice, for example...

    25. Re:FINALLY! by Anonymous Coward · · Score: 1

      AMD still fails to chipset, and still fails to make drivers that are easy to deal with in an enterprise setting. Fix that, and they might have a chance against Intel.

    26. Re:FINALLY! by Anonymous Coward · · Score: 0

      "Anyone who chooses differently than me is spineless!"

    27. Re:FINALLY! by barbariccow · · Score: 1

      openmp?

    28. Re:FINALLY! by Anonymous Coward · · Score: 0

      You think this is madness?

    29. Re:FINALLY! by Anonymous Coward · · Score: 0

      um.. that's how the market's supposed to work.

    30. Re:FINALLY! by epyT-R · · Score: 1

      Sure, if you like hotter/slower.

    31. Re: FINALLY! by Anonymous Coward · · Score: 0

      Phenom II x4 BE @4.0Ghz with 8Gb DDR3 1600 and a 2Gb 7850 Ghz edition still running all new titles (after some updates cough....gta5/FC4...cough....) at 1080p 30fps~ no problems.

      Can't wait to upgrade this year.

    32. Re:FINALLY! by Kjella · · Score: 1

      Well that's a post-hoc justification, if AMD can't compete in a market they can:

      a) Make a comeback
      b) Exit that market
      c) Fail as a company

      If it's a market full of competition b) and c) aren't a big deal but if it's the last competitor and it'll become a monopoly it's a pretty big deal. You can still 'vote with your wallet.' but in a one-party state it's not worth much. A boxer on the ropes doesn't need a knock-out punch to know he's in trouble. It's obvious to everyone, including themselves. And AMD has been diversifying into other markets and dancing on the ropes for quite some time now. Consider these two scenarios:

      AMD Intel
      (Bulldozer) (Sandy Bridge)
      *buy Intel, AMD exits high end market*
      (no offer) (Ivy Bridge - near monopoly rent, little innovation)
      *buy Intel, no real choice*
      (no offer) (Broadwell - near monopoly rent, little innovation)
      *buy Intel, no real choice*
      (no offer) (Haswell - near monopoly rent, little innovation)
      *buy Intel, no real choice*
      (no offer) (Skylake - - near monopoly rent, little innovation)
      *buy Intel, no real choice*
      (no offer) (Kaby Lake - near monopoly rent, little innovation)
      *buy Intel, no real choice*

      AMD Intel
      (Bulldozer) (Sandy Bridge)
      *prop up AMD by buying inferior offer*
      (poor offering) (better offering)
      *prop up AMD by buying inferior offer*
      (poor offering) (better offering)
      *prop up AMD by buying inferior offer*
      (poor offering) (better offering)
      *prop up AMD by buying inferior offer*
      (poor offering) (better offering)
      *prop up AMD by buying inferior offer*
      (poor offering) (better offering)
      *prop up AMD by buying inferior offer*

      Would we be better off in the long run? I'd argue that quite possibly both AMD and Intel customers would be better off in the long run by occasionally taking one for the team, even if AMD customers got the short end of the stick every time. Except we're not a team, so we all do what's best for us individually and lose as a team of consumers. This is not the Intel/Pentium IV situation, when you kick the big incumbent to innovate that's entirely different. Like you, I'm cautiously optimistic that this is AMD's Hail Mary save in the last moment. But it was far from given than this would be the outcome.

      --
      Live today, because you never know what tomorrow brings
    33. Re:FINALLY! by gravewax · · Score: 1

      As much as I want to see strong competition from AMD I will wait for independent verification, we have seen such claims from AMD before turn out to be pure hype and specialised benchmark scenarios that don't represent real world. and how about linking more reputable sites not the steaming turd that is hothardware

    34. Re:FINALLY! by Immerman · · Score: 2

      >I don't care about the power used, I care about the noise needed to dissipate the heat generated

      Umm... power used and heat dissipated are basically identical. The actual twiddling of bits does approximately zero work (as defined in physics), so essentially 100% of the power consumed by a CPU is converted into heat.

      *(there's a theoretical limit, but it's many orders of magnitude less than anything built by humanity)

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    35. Re: FINALLY! by oh_my_080980980 · · Score: 0

      Because there won't be motherboards......

    36. Re:FINALLY! by sexconker · · Score: 2

      Beware the hype train - And this is a hype train of the strongest degree. We're approaching No Man's Sky levels of hype here.

      Keep in mind this is a soft launch. No launch silicon in reviewers hands.No objective reviews. Only "benchmarks" that came from AMD, which will always paint their product in a good light.

      Intel isn't "stagnating" - They're responding to market forces. You really have not needed a faster CPU since the launch of Sandy bridge. What the market wants is more features, lower power. That is what Intel has been focusing on and delivering pretty well.

      Lastly, remember that single thread performance is still king in the desktop space. Lots of cores are great for servers and some applications - But for user facing applications (Including games) the most heavily weighted cpu performance bottleneck is the top speed of your core(s). - Multi-threaded programming is hard and there is no magical compiler switch or library that will suddenly make lots of slow cores = One fast core. (For most applications)

      We all want AMD to give Intel some competition and to push prices down.. But don't hold your breath until reviewers have shipping parts in their hands.

      Reviewers have parts. Parts are up for preorder. Parts are going to be available worldwide on the same date. All signs point to it not being a "soft launch" or a "paper launch". We've seen photos over the past few weeks of people receiving trays of parts. Ryzen looks like it'll be out in volume.

      The AMD-provided benchmarks are objective. They're showing multithreaded performance in a highly-multithreaded workload. They also show one single-threaded benchmark. Yes, the R7 1800X will lose out in single threaded performance against clock-for-clock Kaby Lake and Skylake parts.

      Gamers should look at the Kaby Lake 7700k and the Ryzen 1700X.
      People with highly-multithreaded workloads should look at the Skylake 6900k and 6850k, and the Ryzen 1800X and 1700X.
      People concerned mainly with single-threaded performance shouldn't be looking at any recent part from anybody. For most people, the newer fab processes are simply too tight to allow for clocks high enough to justify replacing shit like the Sandybridge CPUs that have been running at 4.5 GHz - 4.8 GHz for 6 years.

      Intel hasn't been doing SHIT for the desktop CPU market for the past 3 years. Look at http://ark.intel.com/#@Process.... There are a total of 16 SKUs in the i7/extreme class in the last 3 years of the Core products across gens 5-7. 4th gen i7 alone had about triple the number of SKUs. Today Intel shits out a few desktop SKUs and abandons the platform. Kaby lake doesn't even have its full stack out and they're already telling people to wait for Cannon Lake because the value proposition of Kaby Lake or Skylake vs Ryzen is a joke.

    37. Re: FINALLY! by Anonymous Coward · · Score: 1, Informative

      AMD showcased 80+ motherboards available from partners for launch.

    38. Re:FINALLY! by Cederic · · Score: 1

      Ironically the past decade or so is when I've been consciously choosing Intel ahead of AMD.

      Power versus performance, the Athlon 64 was an easy choice but Intel's Core 2 Duo raised the bar again and AMD didn't respond. Since then Intel have set the pace until (hopefully) this one.

    39. Re:FINALLY! by Cederic · · Score: 1

      I know. But it's not the power that deters me, it's the heat dissipation need that deters me. That the two are correlated is just physics.

    40. Re:FINALLY! by Carewolf · · Score: 1

      People with highly-multithreaded workloads should look at the Skylake 6900k and 6850k, and the Ryzen 1800X and 1700X.

      The 6900k and 6850k are not Skylakes, they are Broadwells. Yes, I know they have 6xxx numbers, but Intel adds one to the first number for CPUs with more than 4 cores.. Or something, maybe they give out numbers and brands by dice rolls, who knows..

    41. Re:FINALLY! by aphelion_rock · · Score: 1

      but if it's the last competitor and it'll become a monopoly it's a pretty big deal

      I will be voting with my wallet - I clearly remember back in the 486 days when Intel had the monopoly and were charging more than $1000 for a processor chip, because they could. They were worth more per ounce than gold was at the time.

      We need to maintain competition - Well done AMD, I hope your product lives up to expectations.

    42. Re:FINALLY! by sexconker · · Score: 1

      Broadwell-E, actually. Socket 2011(-3) vs 1151. Switching to "Nth generation" and then putting shit into generations and model numbers they don't belong in is fucking retarded. Soon we'll have Skylake-E, Coffee Lake, and Cannon Lake.

    43. Re:FINALLY! by Immerman · · Score: 1

      "just physics". Thanks, I needed a chuckle.

      Seriously though, it sounds like your issue isn't even the heat, but the noise (and I understand completely). Which suggests perhaps you should focus on quieter cooling systems. At the extreme high end they have completely silent cooling systems designed for use in recording booths, etc. - no fans at all, just a big honking heatsink that doubles as a computer case. And if some sound is acceptable there are lots of less extreme solutions, even $50-1$00 can get you a dramatically quieter cooler than the stock fan and heatsink combo. Especially now that power consumption has been on the decline, so that older solutions can be "underclocked" to be even quieter while providing plenty of cooling for today's less power-hungry processors.

      --
      --- Most topics have many sides worth arguing, allow me to take one opposite you.
    44. Re:FINALLY! by Chalnoth · · Score: 1

      Provided independent benchmarks across a wide range of applications corroborate these results, that is.

      I seriously hope that this finally puts AMD back in the game when it comes to high-performance CPU's. They've been too far behind for too long.

    45. Re:FINALLY! by Chalnoth · · Score: 1

      Also: reliability is an important issue. Intel has traditionally had a pretty significant reliability advantage compared to AMD.

    46. Re:FINALLY! by Hamsterdan · · Score: 1

      Leaving 100$ more for the GPU, which will make a big difference in games.

      --
      I've got better things to do tonight than die.
    47. Re:FINALLY! by Cederic · · Score: 1

      Well, my answer is big fans (so greater airflow at any given RPM) designed to minimise noise (blade shape/angle, bearings) on shock absorbing mounts (minimising the amount vibration is transmitted to the case) controlled intelligently based on motherboard and CPU temperatures.

      Each fan has its own ramp up/down profile, the case is insulated, the closed system watercooling on the CPU has a very very quiet pump and although at full pelt the case can probably take off and hover, under normal and even normal gaming use I can't hear a thing from three feet away.

      When it's 30C outside (and I don't have aircon) and I'm stressing the circuitry for extended periods the fans have the capacity and ability to keep the system well within operational temperatures and I do get some noise, but it's nice being able to cope with that abnormal scenario and still have a silent system the rest of the time.

      It does come at a cost, but that comes to around 5-7% of the total system price. I could understand people preferring to save that cash, I find it valuable.

    48. Re: FINALLY! by Black+LED · · Score: 1

      My first x86 PC had an Intel 8086 that could turbo to 8MHz too.

    49. Re: FINALLY! by Black+LED · · Score: 1

      The K6 came out in 1997 and competed with the Pentium II. Later the K6-2 (and 3) competed with the Pentium III. The AMD CPUs were about on par in integer performance, but got left in the dust when it came to floating point performance. AMD didn't pull ahead of Intel until the Athlon, which was partially due to Intel's poor design for the Pentium 4 but also because the Athlon was a great CPU itself.

      One thing that helped AMD out in the late 90s was 3DNow!, which was introduced in the K6-2 and that quite a few games supported. It really made a noticeable difference if you had decent 3D hardware. At the time, I was running a K6-2 400MHz with dual Voodoo 2 cards in SLI, piped through a Matrox Millennium II. I could run Unreal at 1024x768 with maxed out settings. I remember my roommate and friends being quite envious, heh.

    50. Re:FINALLY! by aliquis · · Score: 1

      Sure.
      But by then we could likely accept the 6 core version to..
      Maybe even the 4C/8T one.

      Or if you are very short on money and don't plan on buying a very capable graphics card anyway the G4560.

    51. Re: FINALLY! by Anonymous Coward · · Score: 0

      lol @ 1080p and 30fps

      Meanwhile my i7 6950X and GTX 1080 runs everything at 4K and 60fps+ at maximum settings.

    52. Re:FINALLY! by Hamsterdan · · Score: 1

      So basically about the same as an i3... nice price. For browsing and everyday use, plenty enough.

      --
      I've got better things to do tonight than die.
    53. Re: FINALLY! by Shirley+Marquez · · Score: 1

      The original AMD 8088 goes back to the days when it was routine for manufacturers of electronic products to insist on second sources being available. It was a true second source of Intel's chip, based on the same die design but made in different foundries. AMD's 80286 was also based on Intel's design. Intel then decided not to share later designs with AMD, so all their later x86 chips were at least partly designed by AMD.

      AMD's first fully in-house design was the K5, but it and the follow-on K6 preserved pin compatibility with the original Pentium. The K7 (Athlon) was the first one to diverge farther from Intel's chips; it was not pin-compatible and required its own motherboard design. The K8 (Athlon 64) was the first chip with the x86-64 instruction set extension that was later also adopted by Intel.

    54. Re:FINALLY! by cas2000 · · Score: 1

      Also agree. I've thought on several occasions over the last few years that it might be worth switching to Intel but every time I've crunched the numbers on specs & pricing, it works out to spending nearly $1000 AUD (for new CPU, motherboard, and RAM) to get performance not much better than what I get now from my collection of Phenom II 1090T and FX-8xx0 machines. All of which are more than adequate for any task I throw at them, from compiling software to running LAN and internet-facing services to watching vids, listening to music, playing games and web browsing.

      It's not even worth it to me to upgrade to current-gen AMD CPUs on most of my machines. My main desktop is a 6-core AMD 1090T (on an Asus Sabertooth 990FX m/b). I would get a noticable benefit from upgrading it to an 8-core FX-8320 or FX-8350, but it isn't *enough* of a benefit to be worth even the ~ $225 AUD it would cost....so spending around $1000 for a similar benefit would be completely insane.

      For games, the GPU counts for far more than the CPU. A 6-core 1090T with a GTX-970 runs every game I have at 1440p at maximum or near-maximum settings, including the latest games released within the last few months. This will be good enough for me until 1070 or 1080 GPUs or better are cheap enough to be worth the price of upgrading from my 970....as a rule of thumb, for a GPU upgrade I want at least double the performance for, preferably, under $300 - which is why I stuck with my GTX-560Ti for years until the 970 was cheap enough.

      Over the last few years, the *ONLY* thing which Intel has that I really want is DDR4 RAM. It's significantly cheaper than DDR3 now, and readily & cheaply available in larger sizes - the most I can get on DDR3 at a reasonable price is 32GB. While 16GB is more than enough on my win7 gaming machine (used solely as a games "console"), for my main desktop and on my servers (all linux), I'd like to upgrade from 32GB to 64GB or more - firefox and especially chromium are both RAM pigs, and file-servers always benefit from more RAM for caching.

      (I usually run both chromium and firefox simultaneously for weeks or months at a time, each with a dozen or three windows open and each windowing having anywhere from 2 or 3 up to 30 or 40 tabs. This consumes a shitload of RAM. After a fairly ruthless tab & window cull this morning, chromium is using about 9GB RAM, and firefox about 4GB. Saving session state and restarting both would probably halve that, they both "leak" memory. They would both be many times worse AND consume far more CPU cycles if I didn't run both uBlock Origin and uMatrix to block ads and javascript)

      PCI-e 3.0 would also be nice, but aside from nvme SSDs there really aren't any devices that benefit significantly from pci-e 3 over pci-e 2. The bandwidth improvement is huge and real but almost nothing actually uses it.

      I'm happy that AMD now has a PCI-e 3 and DDR4 chipset. I'll probably upgrade to Ryzen later this year or early next year after the guinea-pig early adopters willing to pay the must-have-it-now price premium for first release gear have discovered any bugs, but I'm still in no great hurry to upgrade.

      Upgrading to Ryzen will be almost as expensive as switching to mid-range Intel, but a) i'll get a significant upgrade in performance for a price that would get me, at best, a mediocre performance increase with Intel and b) based on past history with AMD chipsets, I'll have an upgrade path for many years that won't require me to toss out motherboard and RAM for almost every new CPU model that comes out.

      Before the current FX-era doldrums that AMD have been stuck in for years, I used to upgrade my CPUs every 2 or 3 years, sometimes more often on particularly good CPU releases. I hope AMD makes it possible to get back into the same upgrade pattern with the new Ryzen gear.

    55. Re: FINALLY! by Bing+Tsher+E · · Score: 1

      An AT&T 6300 (made by Olivetti)? That's the only 8086 based PC clone I can think of off-the-cuff. There were relatively few PC clones made that used the 8086 instead of the 8088. Having to incorporate the BIOS rom in a 16 bit wide data path, for one thing, was more costly than it was worth. 16 bit ROMs are exotic and expensive; two 8 bit ROMs was also expensive. It just wasn't worth it for the slight benefit of the 16 bit data bus at the time, and the first gen ISA bus meant there was no advantage for add on cards.

  2. we've been stuck at 4 core for too long by Anonymous Coward · · Score: 5, Insightful

    Intel has had >4 core CPUs but the affordable stuff for consumers has all been 4 core / 8 thread with the rest of the die given over to GPUs that nobody who needs high performance graphics wants anyway.

    I'd be nice to see AMD back in the game to provide some competition for Intel. Lots of workloads can benefit from more cores: compilation, video processing, simulations, many kinds of "embarrassingly parallel" tasks. Anything you might do with xargs -P.

    If AMD supplies some competitive pressure to push larger core counts down into the affordable price ranges for average buyers, that'll be a good thing. It's been an artificial restriction anyway. Plus it is good for the health of the market to have competition.

    1. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 1

      I'd be nice to see AMD back in the game to provide some competition for Intel. Lots of workloads can benefit from more cores: compilation, video processing, simulations, many kinds of "embarrassingly parallel" tasks. Anything you might do with xargs -P.

      Agreed, but my concern is that they appear to be limiting motherboards to dual channel / 4 dimm configuration. Can any of these new boards do 128GB?

      Virtualization weenies wanna know...

    2. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Some specialized applicatons would benefit from more cores, but these are things that are best done in the cloud and most everyday tasks are linear in nature and do not benefit from more cores. More cores introduce massive power drains and speed hits as data is fragmented between cores for processing and reassembled into final result. There are certain scientific and data processing applications that are suited but these are best left ot server farms and for most gaming and everyday office application needs more cores are a solution that is looking for a problem.

    3. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Sounds like you are buying the wrong parts.

      You do know that Intel makes 6-core parts that don't have the GPU, right?

      Don't just buy the package special from Newegg.

    4. Re:we've been stuck at 4 core for too long by __aaclcg7560 · · Score: 1

      [...] the rest of the die given over to GPUs that nobody who needs high performance graphics wants anyway.

      I wonder if AMD will allow applications to use the GPUs for their own purposes. When I click on the Nvidia applet in my Windows icon tray, it tells me which applications are using the Nvidia GPU.

    5. Re:we've been stuck at 4 core for too long by Joe_Dragon · · Score: 1

      last gen, more workstation based , and caped pci-e lanes.

    6. Re:we've been stuck at 4 core for too long by MightyYar · · Score: 2

      There are certain scientific and data processing applications

      Amazing, that's exactly what I need them for! More the merrier. If you see me on Facebook, it's because I'm waiting on my PC to do something. I use "the cloud" for jobs that are worth the effort of setting up there. But most of the time I'm waiting on my local PC, and double the number of cores would approximately halve that wait.

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    7. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 2, Insightful

      Some specialized applicatons would benefit from more cores, but these are things that are best done in the cloud and most everyday tasks are linear in nature and do not benefit from more cores.

      No... they are very often not "best done in the cloud", because that would mean uploading a few TB of data over my home internet connection to a cloud server. That would take far longer than is practical not to mention bust my ISP cap.

      There are a shitton of common tasks that benefit from having more cores. Hell, just compiling a large project benefits from having a lot of cores. Video transcoding. Image processing. Batch processing of almost any kind. Many kinds of scientific computing. Just because your use of computers is limited to checking your Facebook feed doesn't mean that's the case for everybody. Some of us use computers as computers, rather than as social media portals.

      I can use as many cores as I can get and as the rest of the system (mem bandwidth depending on nature of task, etc) can feed.

    8. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      i7 is still overkill for gaming, and I'm not sure you can get an unlocked i5 without integrated graphics. My current gaming machine has an unlocked i5 in it with wasted GPU silicon, and that's not because I overlooked a part with no GPU.

    9. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      You may have been stuck at 4 cores for too long, but I've been running with 8 cores for several years. AMD very clearly states that my FX-8350 is an 8-core processor. Windows and Linux both report it as having eight cores, too. It might not be as fast as Intel's fastest offering from the same time, but the price/performance made the FX-8350 a much better choice for me.

    10. Re:we've been stuck at 4 core for too long by dunkelfalke · · Score: 4, Interesting

      Old Xeons are, in fact, far cheaper than modern Intel consumer CPUs with a similar performance. Take E5-2670 for example. It costs around 100 bucks, has 8 cores/16 threads and 20 megabytes of cache. For the same money you'll get a Core i3-7100 at best, which has a somewhat better single core performance, but is utterly outclassed in multicore.

      --
      "It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
    11. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Yep. You can save hundreds of dollars going with an AMD MoBo/CPU, dump the savings into a GPU, and it's going to beat an Intel build at the same price point.

    12. Re:we've been stuck at 4 core for too long by thegarbz · · Score: 1

      that nobody who needs high performance graphics wants anyway.

      So ... most people then?

    13. Re:we've been stuck at 4 core for too long by Moof123 · · Score: 1

      There is a chicken and egg argument as well. Many games plateau at 4 cores, but that is no surprise. With 90+% of the installed base being =4 core machines, that is where game studios are going to target. Hopefully with a heap of 6 and 8 real core CPU's in the field there will be a push to take advantage of them.

      In my day job I am a EE. I run a lot of SPICE, FEM, and other scientific apps. We have seen some surprising results with more cores. CST EMS (an electromagnetic simulation tool) claimed it scaled well, and we were hurting big time. We went from running locally on 6 core work stations to a server with a total of 16 cores, and it got SLOWER. WTF? We dug in, and the idiots who wrote the thing had put a tone of effort into making just 1 step of the solving process scalable. So 20% of the time went ~2x faster. The clock speed was about 25% less, so the other 80% of the time it was slower, netting a 15% slower speed. Similarly our headquarters had setup a compute farm to expedite faster solves, but the damn program transferred so much data back and forth (BIG meshes and such, that I could live without, easily) that it was SLOWER to solve on 8 machines than to just run it locally. We eventually ditched them and went to a competitor that was less of a flaming pile of garbage and we've been happy since.

    14. Re:we've been stuck at 4 core for too long by TechyImmigrant · · Score: 2

      There are certain scientific and data processing applications

      Amazing, that's exactly what I need them for! More the merrier. If you see me on Facebook, it's because I'm waiting on my PC to do something. I use "the cloud" for jobs that are worth the effort of setting up there. But most of the time I'm waiting on my local PC, and double the number of cores would approximately halve that wait.

      Ditto. I'm regularly maxing out local i7s on long simulations and computations. I look at all that GPU area on the die and think it would be nice if they added a few more cores instead. I think there are more of us out there than people think. Not everything is serving crappy XML over crappy REST protocols.

       

      --
      I should use this sig to advertise my book ISBN-13 : 978-1501515132.
    15. Re:we've been stuck at 4 core for too long by arth1 · · Score: 1

      with the rest of the die given over to GPUs that nobody who needs high performance graphics wants anyway.

      That's certainly my biggest gripe with new CPUs. I don't want graphics. Or a kitchen sink. If I wanted a SoC, I'd go with an arm solution.

      I want a pure CPU. One that doesn't need a nuclear power plant next door to drive. One that can run for a decade or more should it need to.

    16. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Except most games are CPU-bound, not GPU-bound, and AMD's current offerings are 40% worse in single-threaded situations. You can buy all the GPU you want, but its going to sit there idle.

    17. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      The OS also reports 8 cores with the hyperthreading on i7 CPUs.

      And AMD claiming the FX-series CPUs are 8-core is subject to debate when resources are actually shared between 4 pairs of cores. Hyperthreading on an i7 isn't really 8 cores, but neither is AMD's design.

    18. Re:we've been stuck at 4 core for too long by pecosdave · · Score: 1

      As far as I'm concerned they never left the game, only lost popularity. I'm running a pile-driver core chip on my stuff at home and it doesn't get saturated on my day to day stuff and is quite speedy on my heavy duty stuff as well. I'm not by modern definition a gamer so I'm not pushing it as hard as I can with Windows on the latest AAA titles, but I do use Linux and 3D games - my limitations seem to revolve around my out of date GeForce 750 Ti.

      I just built a pile-driver core machine for work, it's being used for video editing - the editor originally wanted a Mac but we talked him into a custom Windows machine instead since Adobe actually caters to the Windows side more considering their war with Apple. He's exceedingly happy - in fact he's asked us to remove his older Mac from his desk he likes the Windows/AMD combo we built him so well.

      I'm still a fan of the AMD/Nvida combo, from the old ATI Rage IIc almost always having issues in laptops in the late 90's, to my first Radeon literally smoking after playing Alice for about an hour and half, to the conference room machine where I now work having to run on an older version of the Radeon driver if I want sound over HDMI to work I've never been able to bring myself around to liking ATI/Radeon/AMD graphics with the exception of saying they did great in the Wii and GameCube.

      --
      The preceding post was not a Slashvertisement.
    19. Re:we've been stuck at 4 core for too long by AmiMoJo · · Score: 0

      The problem with older CPUs is the motherboards. You will end up with a second hand one, and the chipset will kinda suck. I know because I have this exact problem.

      I have a 2700k from the same era as your example Xeon. It's a great CPU, still very competitive, more than happy with its performance. Problem is, finding a good motherboard is hard. UEFI was new and buggy. The SATA ports are okay but some SSDs don't seem to get the "shut down" signal and experience an "unexpected" power loss. PCIe is limited to 2.0 and the on-board USB 3.0 chipset isn't brilliant either.

      Windows 10 freezes when you enable Bitlocker on it too. Windows 8.1 is okay, but high DPI scaling is terrible.

      I guess you might be lucky and not have any of these issues affect you, but just keep in mind that the Xeon you list came out in 2012 and is now 5 years old, along with all the chipsets and motherboards for it.

      This new AMD CPU might be my opportunity to upgrade. I'd need to new RAM too though, because everything is DDR4 now. And it would be to sort out compatibility issues, not even to improve performance much.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    20. Re:we've been stuck at 4 core for too long by Cesare+Ferrari · · Score: 1

      My preferred choice of machines from that era would be the HP Z600 and Z620 - dual socket capable machines, taking X5600 series in the case of the Z600, and E5 and E5 v2 processors in the Z620 (v2 only if it's the later BIOS version).

      Here in the UK, the Z600 with dual 6 core processors can be picked up with relative ease, and these work well. The Z620 can be harder to find, but can run with dual 8 core processors (say, E5-2670) or dual 10 core processors (E5-2670 v2). The v2s are still very expensive on the second hand market, so stick with an E5 for now. Both machines take heaps of standard DDR3 registered ram, so it's easy to come by.

      If you want rack mount, get an HP-DL380, say a G7, or a G8. The G7s are very cheap these days, whilst the G8 tend to hold their value a bit more. Again, dual socket boards, so you get lots of CPU and plenty of DDR3 slots in them.

      If you worry that you're going to miss out with the slower ram that these machines use, then you're doing it wrong - get your working set down into L3 cache (20mb or so), that'd be my advice ;-)

    21. Re:we've been stuck at 4 core for too long by avandesande · · Score: 1

      If you are running a single GPU the PCIe count has minimal effect on performance for even the highest performing cards. It will start to choke if you are running SLI. (currently gaming on an old 950 PC)

      --
      love is just extroverted narcissism
    22. Re:we've been stuck at 4 core for too long by epyT-R · · Score: 1

      Most of us building high end machines don't 'just' game with them.

    23. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      but just keep in mind that the Xeon you list came out in 2012 and is now 5 years old

      Oh they humanity!!!

      I still have a 8 yr old system as my primary desktop. It works just fine. Your problem also seems to be more with Winblows than with the chip. Get a better OS.

    24. Re:we've been stuck at 4 core for too long by asavage · · Score: 1

      There are lots of pluses with an older Xenon but it also only accepts slower ram and is 32nm instead of 14nm so will use more power.

    25. Re:we've been stuck at 4 core for too long by AmiMoJo · · Score: 1

      Thanks. I don't play many modern games, maybe the odd simulator. The main issue I have is that I only have one slot left capable of more than 1x, unless I drop the GPU down to 8x as well. That might actually be okay since I don't game much.

      --
      const int one = 65536; (Silvermoon, Texture.cs)
      SJW, n: "Someone I don't like, and by the way I'm a fuckwit" - AC
    26. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      I'm sure. You seem to spend most of your time posting sjw bullshit on forums. maybe you should just get an ipad.

    27. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      That's odd i would have said the exact opposite. I guess our anecdotal evidence doesn't seem to speak for "most of us".

    28. Re:we've been stuck at 4 core for too long by epyT-R · · Score: 1

      Only caring what 'most people' want/need is a stupid race to the bottom.

    29. Re: we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Gaming is for losers. No one successful games. Period. Only children have the urge and time to game.

    30. Re:we've been stuck at 4 core for too long by thegarbz · · Score: 1

      Only caring what 'most people' want/need is a stupid race to the bottom.

      Producing something with economies of scale that other people are free to ignore is only racing to the bottom of production costs. The fact that the GPU is included in my CPU doesn't worry me. I don't use it.

      Except for that one time where I had to RMA my normal GPU and I was able to keep using my computer due to this feature I didn't need. If you expect everything to be bespoke then expect everything to cost more.

    31. Re:we've been stuck at 4 core for too long by thegarbz · · Score: 1

      Take E5-2670 for example. It costs around 100 bucks, has 8 cores/16 threads and 20 megabytes of cache. For the same money you'll get a Core i3-7100 at best, which has a somewhat better single core performance, but is utterly outclassed in multicore.

      And the latter will utterly outclass the Xeon in simple things like copying a picture from a memory card, gaming performance given that modern graphics cards require a PCI-e 3.x or latter compliant motherboard (the most common chipset for that processor doesn't support PCI-e 3.0 and good luck finding a motherboard that does. The crappy and expensive Core i3 will be much faster at any memory intensive applications given it's DDR4 support, it has native HD video encoders and decoders. It has new SSE instructions, better power management, better memory protection, the motherboard will likely have eNVM support and blah blah blah

      The ability for a CPU to do math is an incredibly small part of the package that is built around the CPU. I'll happily take a more expensive and slower modern CPU with a decent chipset and feature list.

    32. Re:we've been stuck at 4 core for too long by guacamole · · Score: 1

      Intel has had >4 core CPUs but the affordable stuff for consumers has all been 4 core / 8 thread with the rest of the die given over to GPUs that nobody who needs high performance graphics wants anyway.

      You're making it sound like AMD hasn't been doing the same. For the past four years, all their new consumer CPUs were quad-core APUs with an integrated GPU, while the GPU-less AMD FX processors have been nearly abandoned.

    33. Re:we've been stuck at 4 core for too long by Aighearach · · Score: 1

      I chose an AMD APU because I don't want the energy consumption of high performance graphics, but I do want the API capability to run 3d stuff sometimes. They sell well for a variety of use cases. Probably the majority of non-gaming systems are like this; the users benefit from 3d capability but not from performance.

      Performance is going to mainly depend on choice of motherboard and what IO technologies it uses for the vast majority of professional end-user systems.

    34. Re:we've been stuck at 4 core for too long by sexconker · · Score: 1

      these are things that are best done in the cloud

      Fuck "the cloud".

    35. Re: we've been stuck at 4 core for too long by sexconker · · Score: 1

      Gaming is for losers. No one successful games. Period. Only children have the urge and time to game.

      As an aging gamer I mostly agree with this troll.

    36. Re:we've been stuck at 4 core for too long by sexconker · · Score: 1

      The bulldozer-series "modules" were like conjoined twins. They were mostly separate cores, but shared one of certain pieces between them.
      How well this performed depended on your workload. 256-bit fp would perform as if it were a 4-core, integer would perform as if it were an 8-core.
      They are far more like 2 complete cores than SMT (Intel's Hyper-Threading, and now AMD's ThreadRipper) are.

    37. Re:we've been stuck at 4 core for too long by sexconker · · Score: 0

      Remember kids: When discussing ARM, SoC means shit on a cracker.

    38. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Sure, those second hand E5v1 CPUs are cheap, but not particularly useful when compatible motherboards aren't.

    39. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      No, it's not.

      I play Mirror's Edge Catalyst on two systems: an i3 with a 1060 and an i7 with a 970. MEC runs noticeably smoother on the i7 despite the fact that the 970 is somewhat slower than the 1060 with less video memory.

    40. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Most of what I work on has a lot of potential concurrency. The biggest blocker for implementing concurrency is the linear thinking of most programmers. I've never understood the issue people have with concurrency. I've always found it easy to understand, even before I knew how it worked. Back when I was about 12 years old, I found out about dual socket 486s and spent several hours thinking "how could multiple CPUs work together without slowing each other down?". After some point, I assumed that the CPUs must work independently of each other, but have some limited special ways to update data between each other in "fixed chunks". It wasn't until I was 20 that I learned about atomic operations. By then, they were already a natural way of my thinking.

      What I find stupid is when I ask about ideas on how to make my threaded software faster on StackOverflow. Most of the time people shoot down my thoughts and say "You can't assume that, you're wrong". Then I research the problem some more and find that some highly skilled master in the industry is doing the exact same thing I wanted to do and it's considered an advanced method and lists the exact same corner cases that I mention in StackOverflow. Really?! WTF? From what I have read from blogs and books on advanced methods of multithreading, the masters tell others to "never" do something because nearly everyone who is not a master will do it wrong. They're te kinds of things you can't just cargo-cult, you need to fully understand what's going on.

    41. Re:we've been stuck at 4 core for too long by lsatenstein · · Score: 1

      Intel has had >4 core CPUs but the affordable stuff for consumers has all been 4 core / 8 thread with the rest of the die given over to GPUs that nobody who needs high performance graphics wants anyway.

      I'd be nice to see AMD back in the game to provide some competition for Intel. Lots of workloads can benefit from more cores: compilation, video processing, simulations, many kinds of "embarrassingly parallel" tasks. Anything you might do with xargs -P.

      If AMD supplies some competitive pressure to push larger core counts down into the affordable price ranges for average buyers, that'll be a good thing. It's been an artificial restriction anyway. Plus it is good for the health of the market to have competition.

      In my dreams I see a dual processor home cpu coming. One processor for disk/network I/O, and the other cpu for video, and calculations (free cell, solitaire, and other games), and of course, the more serious stuff such as browsing the web, facebook, instagram, and Gd nows what else.

      I like to twiddle in Linux doing programming in C, so the Ryzen is something I will be acquiring to upgrade from my dual core 2009 system.
         

      --
      Leslie Satenstein Montreal Quebec Canada
    42. Re:we've been stuck at 4 core for too long by unrtst · · Score: 1

      I've been considering something like those for my next desktop. I don't care about graphics performance or the latest PCIe SSd's, but I'm a bit hesitant as I'm unsure of the power usage, heat, and noise.

      Any chance you can comment on those (especially the noise)? I ask because every (rack mount) server I've ever dealt with was noisy, and I haven't touched any non-rack-mount xeons.

    43. Re:we've been stuck at 4 core for too long by Cesare+Ferrari · · Score: 1

      The Z600 and Z620 are a real joy to use and sit by - they are full of low speed fans, and sensible airflow design, so you can get a dual xeon in an under-desk format and these would be my preference. I've actually got one of each, the Z620 is my normal desktop machine, with dual E5-2670, and 48gb of RAM. It has plenty of CPU and RAM to throw at VMs. The Z600 I have has X5650 processors in it, so it's dual six core, and plenty fast enough.

      The DL 380s, G6/G7/G8 are not silent, and you'd get annoyed if you tried to live with one next to you all day. The fans get loud and blowy, but can be set to throttle back. The iLO stuff allows you to remote power cycle these servers, and check their overall happiness. This means that remote operation is much more manageable. I've got 4 (two G6s and two G7s) in my loft, and they are great, doing both CI duty, and two as a test platform for some performance benchmarking. Power use is around 80-100 watts per server, but this is as I have them throttling back to conserve power as they are for development, and normally one is on, and the others I power up when I need them.

      So if you have space, DL380s are great, but if you want to sit next to it all day, go Z600/Z620.

    44. Re: we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Well good. Those 'successful' types are free to slave in their offices all day, sleeping their lives away in pointless meetings and being cucked by their wives at home. I'll pass.

    45. Re:we've been stuck at 4 core for too long by dunkelfalke · · Score: 1

      Not really. On paper DDR4 seems fast, in reality it is barely faster than DDR3, but even the cheapest xeon motherboards support quad channel RAM and usually have twice as many DIMM slots. Also many benchmarks show that there is little speed difference between PCIe 3.0, 2.0 and 1.0 when it comes to gaming, because everything gets preloaded to video RAM anyway. So indeed blah blah blah it is. About the only thing you've got right is the power consumption.

      --
      "It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
    46. Re:we've been stuck at 4 core for too long by dunkelfalke · · Score: 1

      Servers are noisy because their heatsinks and fans are tiny. With a decent heatsink that can dissipate 130w you can operate that old xeon without a fan, or, if you want to be sure, with a virtually silent 140mm fan.

      --
      "It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
    47. Re:we've been stuck at 4 core for too long by dunkelfalke · · Score: 1

      Been lucky so far - got a new motherboard on aliexpress, no compatibility problems with Windows 10, SSDs or gaming whatsoever. For the next 3-4 years I'm good, and then we'll see, but I can imagine going for older server hardware again. It is seriously cheap, especially second hand.

      --
      "It's such a fine line between stupid and clever" -- David St. Hubbins, Spinal Tap
    48. Re:we've been stuck at 4 core for too long by unrtst · · Score: 1

      Thank you! The z620 doesn't look too bad at all. It's going for a bit more money than I expected for a refurb, but it's not very old either (released 3 years ago?). I love the prospect of getting 96gb RAM on a desktop, but I do wish it had more than 3 internal 3.5" drive bays.

    49. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      Some incorrect statements:

      - Modern GFX cards like the Nvidia 1080 or latest Titan Pascal GPU are compatible with but do NOT require PCI 3.0 (PCI is both backward and forward compatible)
      - DDR4 is actually slower than DDR3 even at substantially faster clock speeds (well documented)
      - Hardware video decoding and encoding is almost completely useless on these CPUs. It makes a difference on the i3, sure (2 cores vs 16...)
      - Power management on these CPUs is more or less on par with more modern ones. In fact due to electrical leaks and variability in the production process, the power consumption of latest-generation Intel chips can actually go up compared to previous generations.
      - Very little software today takes advantage of the newer SSE instructions. the E5-2670 is a 2012-era CPU, it is pretty stacked in this department already.

      The ability of a CPU to do math is incredibly important to some users, for instance, but not limited to, people who need or like do math with their computers...

    50. Re:we've been stuck at 4 core for too long by thegarbz · · Score: 1

      And the USB 3 and the onboard encoder and the benefit of the graphics card on chip and the new security features. While we're at it why not talk driver support since there's far newer equipment that doesn't have drivers available so who knows what you end up with.

      Nope thanks. You can have your old xeon, I'll just buy something modern which works.

    51. Re:we've been stuck at 4 core for too long by Anonymous Coward · · Score: 0

      DDR4 is about 1-3% slower 80% of the time, but upto 10% faster 1% of the time, clock of clock, than DDR3. Except it consumes about 50% less power at the same clock and clocks about 100% higher within spec. 2133DDR4 consumes about 20% less power than 1066DDR3. Yeah, you can get your 2133DDR3, but it runs really hot.

  3. Competition Rules by Anonymous Coward · · Score: 0

    This is awesome for computer nerds and builders everywhere. Please, anybody and everybody who can possibly fit AMD products into your plans please do so. I am not a fanboy, but it's crucial that we keep competition alive. It's even good for Intel fanboys because it lowers the prices and increases the performance of their chips too. Everybody wins with competition. Let's keep it that way. At least now it makes some sense from a financial and performance perspective.

    1. Re:Competition Rules by Anonymous Coward · · Score: 0

      Sry, Competition is being deported with the new executive order.

  4. Yay, underdog! by Bing+Tsher+E · · Score: 1

    Let's revel in the underdoggedness.

  5. Re:frist ps0t by mdm-adph · · Score: 0

    Wait folks, don't scold him -- his alternative facts show that he actually *did* have the first post.

    --
    It is by my will alone my thoughts acquire motion; it is by the juice of the coffee bean that the thoughts acquire speed
  6. by 10 full seconds by BKDotCom · · Score: 1

    "converting a 4K 60 FPS video down to 1080P and the Ryzen CPU outpaces the Core i7 by 10 full seconds"

    That's stat a) seems insignificant and b) has zero context
    how about a percentage?
    what was the total time?
    1 sec vs 11 sec, or 3600sec vs 3610 sec?

    1. Re: by 10 full seconds by Bing+Tsher+E · · Score: 1

      It was a 15 hour video.

    2. Re:by 10 full seconds by aliquis · · Score: 2

      In the video Linus did https://www.youtube.com/watch?... the difference was 99.xx vs 112.yy seconds.

      But I don't think that one used a 7700K but a 6800K.

    3. Re:by 10 full seconds by K.+S.+Kyosuke · · Score: 1

      I think it was 91 vs. 112 seconds...

      --
      Ezekiel 23:20
    4. Re:by 10 full seconds by aliquis · · Score: 1

      https://youtu.be/3rUndzpdo1I?t...
      Blender + Handbrake
      1700X $399 vs 6800K $430+ on Intel ARK.
      91.6 vs 112.1

      You are correct. My listening comprehension especially on English numbers aren't the best especially when it's use names like tens (78 hundred for instance.)

      20.5 seconds lower, 112.1*3/4 = 84.1, 84.1/91.6 = 8% faster if viewed as per core and if 8 core Broadwell-E would have had the same clock-rate as 6 core Broadwell-E, which realistically wouldn't had been the case.

      So a very similar performance per core and clock on Ryzen vs Intel Broadwell-E and Skylake though with more cores and threads at the same or lower price than the Intel chips. Better for multi-threaded tasks, possibly worse for single-threaded.

  7. Oh thank god. by Anonymous Coward · · Score: 0

    Intel is in dire need of viable competition. CPU prices are obscene.

  8. About time. by Jethro · · Score: 5, Insightful

    I built myself a gaming PC about two years ago. I've been an AMD supported for decades, so I went with the best CPU AMD was offering at the time. Two years later, it's still the best CPU AMD offers.

    I remember the heyday where AMD actually overtook Intel. Their CPUs were actually better and cheaper. That's no longer the case, but (at least when I built my gaming rig) I was not willing to pay 50%+ more for maybe 10% higher performance, so it was still AMD for me.

    The important thing, though, is that we need competition in order to spur innovation. Before AMD started nipping at Intel's heels, it was all about the MHz (and who could get to GHz first). After that, we started seeing CPUs with more cores and better threading and all the good stuff. I hope Ryzen makes Intel very worried - worried enough that they innovate the hell out of their CPUs. I also hope Ryzen makes AMD enough money that they can continue to innovate, and continue to compete with Intel. Because when that happens, it is we the consumers who win.

    --


    In the land of the blind, the one-eyed man is kinky.
    1. Re:About time. by Anonymous Coward · · Score: 0

      10% more performance?
      Sorry but you don't have idea what you talking about, AMD is not even close to what Intel offers right now.

    2. Re:About time. by Jethro · · Score: 1

      I'm talking about 2 years ago, plus I'm talking about price/performance for a specific point, really, Either way, the price/performance thing is not linear - 50% higher price won't give you 50% higher performance.

      --


      In the land of the blind, the one-eyed man is kinky.
    3. Re:About time. by Anonymous Coward · · Score: 0

      I agree. The last time -- several years ago so take this with a grain of salt -- I went with AMD to build my gaming machine in order to save costs. In the end, it ended up being a complete disaster. Crashed all the time in both the HW and SW realms so I spent way too much time fixing the computer where the costs -- in terms of time -- did not pay out in the end.

      Since then, I've stuck with Intel and have had rock solid performance ever since.

    4. Re:About time. by Gr8Apes · · Score: 2

      I built myself a gaming PC about two years ago. I've been an AMD supported for decades, so I went with the best CPU AMD was offering at the time. Two years later, it's still the best CPU AMD offers.

      Well, I built a PC 7 years ago. It STILL is within roughly 30% of the top comparable system you can build today. These are Intel CPUs btw. Their performance plateaued with the Sandy Bridge core design, although the Gulftown I have competes well with the top end available with even the current i7s. Yes, the 4790K is faster single threaded, but if you OC both the extra overhead available on the Gulftown closes the gap considerably. And overall performance the Gulftown doesn't get doubled until you go to the current 20 core monster that isn't even available for desktop use yet.

      So just realize that Moore's law is done, at least for the last 8 years it has not held. Real performance has barely doubled in the last 7 years by any measure. We have reduced power and heat during that time frame for equivalent performance. So there's still progress, but it's not the 1990s through 2010 heyday we experienced where major performance improvements occurred pretty much every 2 years.

      --
      The cesspool just got a check and balance.
    5. Re:About time. by Jethro · · Score: 1

      I've only ever built one Intel-based PC in... well, ever since I got a... ugh, was it Cyrus? Something... CPU back in the '90s (: I built an Intel-based desktop about 5 years ago because I wanted to try and Hackintosh it, and at the time AMD couldn't do that.

      I have to admit, that desktop still kicks butt. But it was built to last. No way could it run any current game in any decent quality, even if I stuck a GTX1080 in it.

      --


      In the land of the blind, the one-eyed man is kinky.
    6. Re:About time. by Jethro · · Score: 1

      I hear what you're saying, but I'm not talking about it being close to the same performance. It's literally the same exact CPU model that was the top model 2 years ago. That's some crazy stagnation there.

      But, going back to what you say, that reinforces my point. AMD hasn't done anything in the CPU arena in years. Why would Intel have to innovate? We need a competitor to nip at Intel's heels and get them moving again, and ideally it'd be great if Intel and AMD can race each other for supremacy forever because then we get constant innovation.

      --


      In the land of the blind, the one-eyed man is kinky.
    7. Re:About time. by PIBM · · Score: 1

      A lot of people are comparing the CPU price; if I drop +50% on the CPU I barely get +10% better performances! But they fail to take into account the fact that they are dropping 3 grand on the system, so that 200 extra bucks only represent a 6% premium for 10% better performances. I expect that`s what you did too.

    8. Re:About time. by Solandri · · Score: 1

      I remember the heyday where AMD actually overtook Intel. Their CPUs were actually better and cheaper.

      Not cheaper by much. I remember comments here asking why if AMD had overtaken Intel, why were their high-end CPUs now almost as expensive as Intel's had been. The replies stating that if you're the market leader, you can set your prices as high as you want.

      I think that's where they screwed up. Instead of keeping their prices low so they could gain market share, they raised their prices to try to recoup their R&D expenses. If they'd aimed for market share, that would've resulted in much greater industry support - AMD motherboards, chipsets, compatible memory, retailers stocking their CPUs, brand names offering more models using AMD processors, etc. Instead they chose short-term gains over long-term, meaning Intel had an easy time reasserting its dominance as soon as they managed to convert their laptop CPU cores to desktop use.

      If they're keeping the price down at half Intel's prices, it sounds like they learned their lesson.

    9. Re:About time. by Anonymous Coward · · Score: 0

      I've done multiple AMD builds in the last several years and had no such problems. I do tend to go with Nvidia GPUs though.

    10. Re:About time. by Anonymous Coward · · Score: 0

      If they can't recoup their R&D expenses, they wouldn't be here.

    11. Re:About time. by Jethro · · Score: 1

      Oh, there were some questionable decisions, to be sure (remember Slot-A CPUs?)

      --


      In the land of the blind, the one-eyed man is kinky.
    12. Re:About time. by Aereus · · Score: 1

      AMD has been shipping approximately the same CPUs for like 5+ years now. Their last architecture flopped and they've been stalled ever since. Last relevant CPU release they had was the FX-8350, and that is from 2012. Even then, the IPC on them was so low, they benched lower than most i3 offerings.

    13. Re:About time. by Jethro · · Score: 1

      Yeah, and that made Intel sit on their laurels and not really bother. I'm hoping this means more competition!

      --


      In the land of the blind, the one-eyed man is kinky.
    14. Re:About time. by Anonymous Coward · · Score: 0

      What? 3 grand on a AMD build... are you for real? In the real world it's probably more like 1 grand, tops.

      You know that already but let me remind you that AMD is used mostly by people on a budget (tight one). A 3 grand build don't fit the target audience.

      I know... it isn't fair because it ultimately breaks your little percentage game. Tough luck.

    15. Re:About time. by PIBM · · Score: 1

      Humm, indeed my friends and family is no longer representative of the kids building cheap system, and they tend to go with at least 1 GTX1080, or used to have 2 980ti, purchase large 4K tv or high quality monitor, or in my case I`m running multiple high resolution monitors. 1 or 2 1TB SSDs, one of them being m2 NVMe 4x, along with a large platter disk (or 2 for raid redundancy), 200$ on keyboard & mouse is a minimum.

      Albeit one thing I`ll give you is that the price I was referring to was in CAD, but the +% should still match.

    16. Re: About time. by Anonymous Coward · · Score: 0

      This has nothing to do with AMD and everything to do with you not being able to build a computer worth a fuck.

    17. Re: About time. by Anonymous Coward · · Score: 0

      If a far girl is wearing a backpack, what is the probability there is food inside the backpack?

    18. Re:About time. by Anonymous Coward · · Score: 0

      Err, Moore's law doesn't have anything to do with speed, it's how many transistors you can pack in to a square inch.
      It's held true up until now, but unlikely to hold going forward...

    19. Re:About time. by AvitarX · · Score: 1

      Didn't they have capacity problems?

      That's part of why spinning off their fabs was a good call.

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
    20. Re:About time. by Anonymous Coward · · Score: 0

      Are you implying that they were somehow quantifiably worse than Intel's Slot 1 CPUs? The first 1GHz processor was Slot A.

    21. Re:About time. by Gr8Apes · · Score: 1

      It hasn't held true, since about 6 years ago. Fabrication processes have slowed down miniaturization, and heat has killed off creating more powerful systems. (both of which moore's law implied. We all knew there would be a plateau reached, we just got there faster than expected. Architectures changed in the early 2000s, and then again in the late 2000s. We are now requiring a new architecture.

      --
      The cesspool just got a check and balance.
    22. Re:About time. by Gr8Apes · · Score: 1

      That's definitely some stagnation, although I thought AMD was pushing out some new CPUs? I'll be honest, when my current CPU hasn't been bested by enough to make an upgrade worth looking at, I haven't really kept up to date on all CPU advancements, or lack thereof. Yay - here's a $1500 part that goes 10% faster... Wake me when it's a $100 upgrade for 10%.

      --
      The cesspool just got a check and balance.
    23. Re:About time. by Jethro · · Score: 1

      Yeah, this article is about the new/upcoming AMD CPUs (:

      --


      In the land of the blind, the one-eyed man is kinky.
    24. Re:About time. by Gr8Apes · · Score: 1

      Until they can double the performance of an OC'd 980x and do it for a total cost of less than $500, I will not be interested. :)

      --
      The cesspool just got a check and balance.
  9. Any one have a block-map as to how the pci-e is by Joe_Dragon · · Score: 1

    Any one have a block-map as to how the pci-e is setup?

    Also how will the server chips be setup in pci-e lanes?

  10. From what I see, AMD's done a GOOD job! by Anonymous Coward · · Score: 0

    See my subject & this article from The Register http://hothardware.com/news/amd-ryzen-7-1700x-thrashes-intel-core-i7-5960x-benchmark-leak/ & specifically/graphically, this test result http://hothardware.com/ContentImages/NewsItem/40273/content/Cinebench-Ryzen-7-1700X-HotHardware.png/

    * If they keep this up, I know what will replace my Intel Core I7 4790k next round (2++ yrs. into it now, another 3 to go most likely before I buy again though).

    APK

    P.S.=> In any event? COMPETITION IS GOOD - it drives up the bar for the competitors (AMD & Intel) & who gains? We as consumers do, via superior choices offered... apk

    1. Re:From what I see, AMD's done a GOOD job! by Anonymous Coward · · Score: 0

      "COMPETITION IS GOOD"

      Says the asshole who shouts down people with far better security solutions with his constant deluge of bullshit.

      Son, you can't even handle competition in your own space. You're not allowed to speak.

      Also, looks like WHIPSLASH FAILED TO GET RID OF APK.

    2. Re:From what I see, AMD's done a GOOD job! by BarbaraHudson · · Score: 1

      Be careful what you ask for. Do you really want slashdot to start censoring posts?

      --
      "Transparent" is a shit show that trades on every stereotype going. A man in drag is NOT a transsexual.
    3. Re:From what I see, AMD's done a GOOD job! by ArchieBunker · · Score: 1

      They have deleted and censored posts in the past. I actually had my account banned one for making comments.

      --
      Only the State obtains its revenue by coercion. - Murray Rothbard
    4. Re:From what I see, AMD's done a GOOD job! by BarbaraHudson · · Score: 1

      You can get a temporary ban (I wonder if it's still the "pink page of death") by having too many down-mods - even with excellent karma (go figure). I wouldn't call that censorship any more than having a comment down-modded to -1. It's just a feature of the system, same as rate-limiting, that has no permanent effect.

      --
      "Transparent" is a shit show that trades on every stereotype going. A man in drag is NOT a transsexual.
  11. Re:frist ps0t by epine · · Score: 3, Funny

    alternative facts

    We used to call them "hot grits". In was an everyday topic of discussion.

  12. More cores by kcdoodle · · Score: 3, Interesting

    I run all my Windows machines Virtually. I really like to be able to dedicate a core or two to each Virtual machine and still have enough left over for my Linux host OS.
    Four cores just ain't enough for me. I'm looking forward to 128 core processors...

    --

    - I live the greatest adventure anyone could possibly desire. - Tosk the Hunted
    1. Re:More cores by Anonymous Coward · · Score: 0

      640 cores should be enough for anyone.

    2. Re:More cores by thegarbz · · Score: 1

      I knew Windows wasn't good at multitasking, but what are you doing? Spinning up a new VM for each thread?

    3. Re:More cores by sl3xd · · Score: 1

      Intel MIC architecture (Xeon Phi) is probably the closest at 72 Xeon cores (and 4 threads/core).

      --
      -- Sometimes you have to turn the lights off in order to see.
    4. Re:More cores by Anonymous Coward · · Score: 0

      This! So Much This!

      Most modern games need more than 2 physical cores. Leaving only one core for the host is painful if you're watching HDTV with integrated graphics or trying to decode / encode something (incl compression). It's somewhat better if there are two logical cores, but having 6+ physical cores in a consumer grade (single socket) motherboard would be best. Anyone using VGA-Passthrough for windows gaming should take heed.

    5. Re:More cores by rrohbeck · · Score: 1

      Naples looks incredibly yummy. Hopefully it'll be launched soon too. If the prices are halfway reasonable I might build with that. Dual socket 32C64T?

  13. and stuck at 16 + DMI pci-e with to much on the by Joe_Dragon · · Score: 1

    and stuck at 16 + DMI pci-e with to much on the DMI link.

    For video X16 or X8 X8 is good for most cards. But putting storage / network / sound / usb / etc on the DMI is overloading it.

    Also Intel caped the $350+ lowend cpu's down from 40 lanes to 28 lanes in boards setup for 40 lanes per cpu making people take an $150+ jump for a small clock boost. Back in Ivy Bridge-E all chips had the same pci-e lanes.

    1. Re:and stuck at 16 + DMI pci-e with to much on the by Anonymous Coward · · Score: 0

      There is zero evidence to support this. You will not saturate DMI in a gaming or consumer workload. Also - Multi-GPU setups are retarded. Even AMD and Nvidia are abandoning this market.

      If you're pushing these work loads pay for Xeon based platform that uses QPI. They're not even that expensive.

    2. Re:and stuck at 16 + DMI pci-e with to much on the by Joe_Dragon · · Score: 1

      QPI is not used for the chipset link.

      and on desktop boards you have storage pci-e X4 (per card), usb 3.X upto X4, networking X1 per gig-e nic, sound x1, other stuff as well. All over an X4 dmi link.

    3. Re:and stuck at 16 + DMI pci-e with to much on the by Anonymous Coward · · Score: 0

      you have storage pci-e X4 (per card), usb 3.X upto X4, networking X1 per gig-e nic, sound x1, other stuff as well. All over an X4 dmi link.

      And it's not really a problem because those things are never pushing data full blast at the same time.

      Hell, sound and networking can't even saturate a single PCIe link.

      If you are in the 1% of users who actually pushes that much data around, chances are you need the extra memory bandwidth and CPU cores available on their high-end workstation platform.

      Just because you want QPI doesn't mean you actually need it.

    4. Re:and stuck at 16 + DMI pci-e with to much on the by Joe_Dragon · · Score: 1

      QPI is only for to cpu to cpu not cpu to chipset.

      Intel has been stuck on DMI + X16 pci-e for way to long.

      AMD is uping Intel on the desktop side. I want to see what the ryzen serer / workstation systems will be like as well.

  14. All of this has happened before... by MachineShedFred · · Score: 5, Insightful

    It seems that every time Intel gets a significant pile of laurels, they like to rest on them. Then someone comes up from behind to kick them in the ass. AMD has done it before, perhaps with this generation they can do it again.

    And who wins? We all do. Last time, Intel got off their ass and created the Core-series that has expanded PC processing power to the point where upgrade cycles have gone from 3 years to 6+. Let's hope that this shot across the bow ushers in a new era of chip design that brings features we want, rather than the features that they think we want.

    --
    Slashdot still doesnâ(TM)t support Unicode after it was added to the HTML standard in 1997.
    1. Re: All of this has happened before... by Anonymous Coward · · Score: 1

      If only something similar would happen with operating systems....

    2. Re: All of this has happened before... by Anonymous Coward · · Score: 1

      Hasn't it happened with operating systems too? I haven't upgraded the OS on my computer in more than 6 years.

    3. Re: All of this has happened before... by Anonymous Coward · · Score: 0

      Intel rest on their laurels as a matter of strategy. They already have the next AMD-killer, but they won't release it until they absolutely have to. They essentially are milking the most value they can, as slowly as the competition lets them. So we still win because of AMD's competition, but remember there is no benefit for Intel to release a new product before they absolutely have to, because they make more money shipping "old" products as the state-of-the-art, vs. shipping their new stuff with more expensive processing and lower yields which forces them to drop the price of the last generation.

    4. Re:All of this has happened before... by Solandri · · Score: 3, Informative

      That's the way it looks to enthusiasts, but that's not what Intel has been doing. About a decade ago, we hit the point where processors were "fast enough" for mainstream tasks. People stopped buying i5 and i7 processors, in favor of i3, Celeron, and Pentium. For the last 10 years, only enthusiasts and gamers have cared about improved performance. The vast majority of the market cared more about power consumption. Intel hasn't been worried about AMD, but they were scared to death of ARM. They rushed to bring Atom to market to keep the low-end on x86/x64, instead of moving to ARM.

      So they haven't been resting on their laurels. They've been working hard at reducing power consumption. That's what really hurt AMD after they lost their performance lead. For a few years AMD still offered more performance per Watt, making AMD the natural choice for moderate-load servers and systems meant to be left on 24/7. But Intel soon beat AMD there, taking away AMD's only advantage. (That's when AMD used their ATI acquisition to integrate a GPU which could beat Intel's integrated GPU - essentially carving out a spot in the low-price gamer market.)

      A Core 2 Duo system would use about 70 Watts idle, 150 Watts under load. A Sandy Bridge system would use about 45 Watts idle, 120 Watts under load. A modern Skylake system uses about 35 Watts idle, 80 Watts under load. Subtract the 20-30 Watts consumed by components other than the CPU, and the reduction in CPU power consumption over the last 10 years has been remarkable.

    5. Re: All of this has happened before... by Waffle+Iron · · Score: 1

      If only something similar would happen with operating systems....

      I'd bet that most people would prefer it if Microsoft were still resting on its Windows 7 laurels.

    6. Re:All of this has happened before... by Anonymous Coward · · Score: 0

      It seems that every time Intel gets a significant pile of laurels, they like to rest on them. Then someone comes up from behind to kick them in the ass.

      Posting anonymous because reasons. But this is more true that you can know. I expect there to be some more hiring in the CPU teams soon.

    7. Re: All of this has happened before... by drakaan · · Score: 1

      Every time I hear people bitch about windows 8/8.1/10, I shake my head. Win7 was good. Heck, it was my favorite version of Windows until 8 came out...it was better than Win2k Pro, but Win10 is better still. It's more stable, has less crufty old bits, less driver problems, less security problems, and generally causes me less headaches. Then again, I'm not most people.

      --
      "Murphy was an optimist" - O'Toole's commentary on Murphy's Law
    8. Re: All of this has happened before... by ChrisMaple · · Score: 2

      Intel rest on their laurels as a matter of strategy.

      If that's what Intel is doing, it's a stupid strategy. If Intel could produce a processor running at 8 GHz but otherwise identical to an I7-7700K, they could charge $3000 apiece for them and sell as many as they could make. Intel doesn't because it can't; physics and present technology don't allow it.

      --
      Contribute to civilization: ari.aynrand.org/donate
    9. Re: All of this has happened before... by DeBaas · · Score: 1

      You're probably right except for the update madness. I sometimes dual boot to Windows 10 on a quad core I7 with 32GB and a ssd drive. All to often at boot I had to wait over half an hour looking at the blue screen telling me to 'not power down update in progress' . This is not exaggeration, over half an hour on boot!
      If there is one area where Linux really shines compares to Windows it's on this. At most my distro will tell me that it needs a reboot for the upgrade to take effect. A normal reboot... no crap as it shutting down, nor when it boots. Just boots with the new kernel.
      Windows 7 had these long wait for update issues too, but it didn't take this long. That's a new nice and shiny Windows 10 feature.

      --
      ---
    10. Re: All of this has happened before... by Anonymous Coward · · Score: 0

      Exactly that "Waiting for the computer to do something that was not initiated by user" in Windows 7 you are almost the driver after that in 8 and 10 you are becoming a passenger :)

    11. Re: All of this has happened before... by Anonymous Coward · · Score: 0

      I haven't upgraded my OS in about the same time. I built my computer with Windows 7, and it's still running great. Better than any Windows 8.1 or 10 computers I've used.

      Referring to two of the comments above...
      "that brings features we want, rather than the features that they think we want."
              "If only something similar would happen with operating systems...."

      I found a Linux distribution that fits that bill nicely. I've been using it on my laptops and computers, except for my gaming rig, since early 2011.

    12. Re: All of this has happened before... by Anonymous Coward · · Score: 1

      But conspiracy theories are more fun to believe

    13. Re: All of this has happened before... by oh_my_080980980 · · Score: 1

      LMOL ok Potsy. Intel has fabrication issues. They can't go smaller based on their current process. They're stuck and can't get to 10nm.

    14. Re:All of this has happened before... by oh_my_080980980 · · Score: 1

      Seriously? Hate to break it too you Potsy but no consumer demanded faster processors. Intel pumped them out. People bought them. That's it. Intel decided to stop pushing the envelope. That had nothing to do with consumers.

    15. Re: All of this has happened before... by Anonymous Coward · · Score: 0

      > Then again, I'm not most people

      No, most people don't feed all their data directly to Microsoft. You, presumably, do so willingly.

    16. Re:All of this has happened before... by aliquis · · Score: 1

      Rumor are Intel will get rid of some legacy stuff ~2019-2020.

    17. Re: All of this has happened before... by aliquis · · Score: 1

      Reminds me I can't have gotten updates for over two weeks.
      Maybe I've finally been able to disable them and should search for some?

    18. Re: All of this has happened before... by Bert64 · · Score: 1

      But those sales wouldn't be in addition to their existing high end processors, they would be instead of, plus with a $3000 price tag the sales would be low and economies of scale wouldn't kick in so the production cost would remain high too.
      Their existing designs are well tried and tested, R&D costs are low and the fabrication process has been refined sufficiently to have very high yields so the margins are very good. There really is no incentive to sell anything significantly better unless forced to by competition.

      The Alpha processor used to be what you describe - massively faster than anything else, but highly expensive. They remained an expensive niche for people who needed the highest performance at any cost.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    19. Re: All of this has happened before... by llZENll · · Score: 2

      Really but why? I've quite enjoyed not having to upgrade my sandy bridge. I overclocked it to 4.5Ghz 6 years ago and it's still faster in everyday tasks than even the latest $1000 offering, only in extreme multi core tasks does it lack, which I rarely need.

      A new system is such a distraction from my work. Days to reinstall and update everything, weeks to weed out issues and acclimate to a totally new setup.

      Now if we could only get Microsoft to support Windows 7 indefinitely. I'd gladly pay $10/year for infinite support, or at least for another 5-10 years before AI takes my job.

    20. Re:All of this has happened before... by Anonymous Coward · · Score: 0

      You are incoherent.

    21. Re:All of this has happened before... by cas2000 · · Score: 1

      I don't care much about power consumption. I don't give a damn whether a CPU is going to cost me $30 for a year's power consumption vs $60 or even $80. It's a trivial difference - the remainder of the system (motherboard, ram, disks, gpu, monitor, etc) will add up to twice as much or more than the CPU anyway. Caring about it makes as much sense as caring about the daily ups and downs of petrol prices....an extra 2c or even 10c per litre makes exactly fuck-all difference over an average 50-80 litre tank.

      I care a lot more about the fan noise, but even that's cheap and easy to mitigate or eliminate.

      What I care about is a) the price, and b) the price:performance ratio. AMD shits all over Intel for both of those. It always has and the Ryzen looks like it's continuing that pattern.

      Sure, Intel has the top end CPUs. But paying 5 times or more the price for a CPU barely twice as fast is just plain stupid unless you're doing something where every cycle of performance really matters (like building a HPC or VM-hosting cluster) and you have a huge budget. The new top end Ryzen CPUs look like being almost as fast as Intel's top end CPU for about half the price.

      BTW, Intel doesn't much care what AMD do. They'll make minor attempts to compete, but the fact is that they need AMD to exist so that they can avoid anti-monopoly lawsuits and regulations. They'll cede the low-end and some of the middle of the market to AMD because the massively over-priced top-end is their cash cow.

    22. Re: All of this has happened before... by drakaan · · Score: 1

      True. Most people feed all their data directly to Google willingly. Of course, I do that, too.

      --
      "Murphy was an optimist" - O'Toole's commentary on Murphy's Law
  15. I'll believe it... by __aaclcg7560 · · Score: 1

    When Newegg has listings for the new processors and motherboard. Although it might be too late for me since I retired my nine-year-old Vista-compatible AMD quad-core motherboard for a Windows 7-compatible AMD eight-core motherboard last year. I might let the platform mature before I spring for new hardware.

    1. Re:I'll believe it... by David_Hart · · Score: 2

      When Newegg has listings for the new processors and motherboard. Although it might be too late for me since I retired my nine-year-old Vista-compatible AMD quad-core motherboard for a Windows 7-compatible AMD eight-core motherboard last year. I might let the platform mature before I spring for new hardware.

      Based on the news articles, you should start seeing these on Newegg within a couple of weeks (March 2nd). Supposedly AMD primed the retail channel prior to the announcement.

    2. Re:I'll believe it... by drinkypoo · · Score: 1

      I might let the platform mature before I spring for new hardware.

      And also let intel bring out some new processors, or make some price drops, and let AMD make some price drops to compensate. Because I, for one, am not spending more than two hundred bucks on a CPU. (I have an FX-8350 right now.)

      Yes, I am a cheap bastard. If I weren't, I would have bought an Intel processor.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
    3. Re:I'll believe it... by MiniMike · · Score: 1

      Newegg has 7 Ryzen motherboards listed. They're posted as 'out of stock' probably until the launch on March 2. Found them simply by searching for 'ryzen'. No CPUs listed yet.

    4. Re:I'll believe it... by __aaclcg7560 · · Score: 1

      Newegg has 7 Ryzen motherboards listed.

      Supposed to be 81 motherboards available at launch. The few that Newegg has listed are all ATX boards. I like to build my systems small with micro-ATX and mini-ITX boards.

    5. Re:I'll believe it... by Moof123 · · Score: 1, Funny

      Aren't you a special little princess.

    6. Re:I'll believe it... by __aaclcg7560 · · Score: 1

      Aren't you a special little princess.

      In what way?

    7. Re:I'll believe it... by MiniMike · · Score: 0

      I like to build my systems small with micro-ATX and mini-ITX boards.

      Well, Amazon doesn't have any Ryzen CPUs or Motherboards, but they do have this Wait for Ryzen t-shirt, perhaps you could order it in small while you wait?

    8. Re:I'll believe it... by Anonymous Coward · · Score: 0

      Amazon WTF?

    9. Re:I'll believe it... by __aaclcg7560 · · Score: 1

      Well, Amazon doesn't have any Ryzen CPUs or Motherboards, but they do have this Wait for Ryzen [amazon.com] t-shirt, perhaps you could order it in small while you wait?

      I fail to understand why building small systems is an issue for some people. I have multiple systems in a limited space. My FreeNAS file server is a full ATX tower with six hard drives and room for another six hard drives. My gaming and Linux boxes are micro-ATX boxes. If I ever build out a Beowulf cluster, it will be four mini-ITX motherboards stacked on top of each other in a custom built, one-cubic-feet case.

    10. Re:I'll believe it... by moeinvt · · Score: 0

      "I fail to understand why building small systems is an issue for some people."

      Big hands.

    11. Re:I'll believe it... by __aaclcg7560 · · Score: 1

      Big hands.

      So what? I got big hands and bad eyes Tweezers and magnifying lens help with those problems..

    12. Re:I'll believe it... by Anonymous Coward · · Score: 0

      Sure they do - they release on the 2nd though. Embargo was supposed to lift the 28th but a leak occurred last night and it sounds liek we should see benchmarks from others soon.

      https://www.amazon.com/AMD-YD180XBCAEWOF-Ryzen-1800X-Processor/dp/B06W9JXK4G

    13. Re:I'll believe it... by wbr1 · · Score: 1

      Listings are up. CPU ships with no cooler, and no coolers yet, but there you go, preorders of CPU and mobo are there. https://www.newegg.com/Product...

      --
      Silence is a state of mime.
    14. Re:I'll believe it... by Anonymous Coward · · Score: 0

      You misspelled 'FTW'!

  16. I fell back in love with them... by Anonymous Coward · · Score: 0

    when their stock popped from $2.50 to $14.30!

    1. Re:I fell back in love with them... by Anonymous Coward · · Score: 1

      I bought the stock at $3 and just sold it at $14. Over the past decade or so, I have done that several times with AMD. I love the company, they have bought me two cars. lol

  17. What about single thread performance? by fred6666 · · Score: 3, Interesting

    It still matters for most people, and it has been a problem for AMD.
    Will they offer lower priced Core i5/i3 competitors based on this architecture?

    1. Re:What about single thread performance? by K.+S.+Kyosuke · · Score: 2

      They will. Apparently slightly later. Maybe they'll even follow the Kaveri/Carrizo/Excavator model and release parts for laptops in a single line with with four-core desktop CPUs. This hasn't been announced but I'd find it perfectly logical since it worked quite well so far.

      --
      Ezekiel 23:20
  18. Paper launch by Guspaz · · Score: 1

    Are they on sale yet? Are there any reviews of it? If the answer to both of those is no, then it hasn't launched.

    1. Re:Paper launch by TimothyHollins · · Score: 0

      Have you not heard of AMD?

      Of course it's a paper launch!

    2. Re:Paper launch by sl3xd · · Score: 1

      And of course the comparison is a paper launch from AMD vs a shipping processor from Intel.

      --
      -- Sometimes you have to turn the lights off in order to see.
    3. Re:Paper launch by Anonymous Coward · · Score: 0

      It's sufficient to read the article. They will be on the shelves from march 2nd 2017 (IOW, in a couple of weeks).

    4. Re:Paper launch by iCEBaLM · · Score: 2

      https://www.youtube.com/watch?...

      You can preorder them on newegg. Release is apparently Thurs/Fri next week.

    5. Re:Paper launch by jimbo · · Score: 2

      Well, with only a week from "launch" to availability it's perhaps unkind to call it a paper launch - those boxes are already on their way to retailers.

    6. Re:Paper launch by Guspaz · · Score: 1

      Then the launch is next week, and the statement that "AMD Launches Ryzen" is false. I should note that it's highly suspicious for a new CPU's review embargo to be on the same day as retail availability.

  19. I don't think many apps use multi core by rsilvergun · · Score: 1

    And besides Ashes of the Singularity I can't think of any that use more that four. Heck, Far Cry 3 only needed four cores because the devs bound to core 3 by mistake. There was a fan patch that forced it to bind to core two and got it running on dual cores. Multi core programing is dammed hard. It hasn't been worth it except for a handful of apps like video encoders...

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
    1. Re:I don't think many apps use multi core by TheFakeTimCook · · Score: 1

      And besides Ashes of the Singularity I can't think of any that use more that four. Heck, Far Cry 3 only needed four cores because the devs bound to core 3 by mistake. There was a fan patch that forced it to bind to core two and got it running on dual cores. Multi core programing is dammed hard. It hasn't been worth it except for a handful of apps like video encoders...

      What's so hard?

      It's just a matter of semaphores and some intelligent decisions on what goes where.

      Hell, Apple already did the hard part for you, and then gave it to the planet for free...

    2. Re:I don't think many apps use multi core by BrendaEM · · Score: 1

      Photoshop, and Premier, Handbrake, ffmpeg, Winff--almost anything that works on the video you see on youtube were likely done with multiple cores. The marketing images of things you buy were likely raytraced on multiple cores. The things in your life were designed with multi-core computing, such as the cars you drive, and the buildings you stand in. To name a few.

      --
      https://www.youtube.com/c/BrendaEM
    3. Re:I don't think many apps use multi core by Anonymous Coward · · Score: 0

      > I can't think of any that use more that four

      That's like someone from the 1800s observing that all non-animal driven vehicles need tracks, because he's never seen one that doesn't. If half the gaming market had 16 core Xeons, I betcha some games would scale a lot better with cores.

    4. Re:I don't think many apps use multi core by sexconker · · Score: 1

      And besides Ashes of the Singularity I can't think of any that use more that four. Heck, Far Cry 3 only needed four cores because the devs bound to core 3 by mistake. There was a fan patch that forced it to bind to core two and got it running on dual cores. Multi core programing is dammed hard. It hasn't been worth it except for a handful of apps like video encoders...

      Did it run on the 3 core Phenoms?

    5. Re: I don't think many apps use multi core by Anonymous Coward · · Score: 0

      Lol. Another arm chair concurrency expert. If he really had experience developing multi threaded applications, he wouldn't say it. Dumb shit again. So many all around. Look at this one.

  20. 20% isn't enough by rsilvergun · · Score: 1

    Though 50% is. But at 20% I'm left with the same problem I had with the 8350. The processor demands a much better motherboard that eats up the savings. Cheap AMD motherboards ruin their performance...

    --
    Hi! I make Firefox Plug-ins. Check 'em out @ https://addons.mozilla.org/en-US/firefox/addon/youtube-mp3-podcaster/
  21. Yet more skewed bullshit testing from AMD. by JustNiz · · Score: 0

    I see AMD are still up to their old tricks of trying to make gross performance statements based solely on hand-picked benchmarks that capitalize on their product having twice as many cores (or in the case of video cards, twice as many GPUs) as the competitions unit that they chose for comparisons.

    I'd like to see a like-for-like benchmark between Ryzen and I7, such as single-thread at the same clock speeds. I suspect AMD won't be publishing that result anytime soon though.

    1. Re:Yet more skewed bullshit testing from AMD. by avandesande · · Score: 1

      It matches the Core i7-6900K core for core and was faster in the first test- can't be that far off.

      --
      love is just extroverted narcissism
    2. Re:Yet more skewed bullshit testing from AMD. by sl3xd · · Score: 1

      You're saying a comparison to a CPU core nearly a year older is a fair comparison?

      The story hasn't changed in nearly a decade: AMD boasts they have "massive" performance gains over their old hardware, and that they can selectively beat Intel's old hardware.

      The problem is that Intel invariably has its own improvements, and they easily beat AMD's best-case hype.

      I really would like to see AMD return to beating Intel's chips, but this doesn't leave me hopeful.

      --
      -- Sometimes you have to turn the lights off in order to see.
    3. Re:Yet more skewed bullshit testing from AMD. by avandesande · · Score: 1

      There is hardly any difference on a per core basis between any intel chips for the last two or three years.

      --
      love is just extroverted narcissism
    4. Re:Yet more skewed bullshit testing from AMD. by iCEBaLM · · Score: 1

      You're saying a comparison to a CPU core nearly a year older is a fair comparison?

      It's not AMDs fault it's the latest and fastest Intel "High End Desktop Processor". The only thing higher end is the 6950X at 10 cores with an MSRP of $1700.

    5. Re:Yet more skewed bullshit testing from AMD. by Areyoukiddingme · · Score: 1

      I'd like to see a like-for-like benchmark between Ryzen and I7, such as single-thread at the same clock speeds.

      Uh, they did. The Cinebench single-threaded results are in the slide. Right hand side. The 1800X is indistinguishable from Intel's 6900K at single-threaded performance. And Cinebench is compiled with Intel's compiler.

      Undoubtedly there will be some benchmarks where Intel is still ahead, and yes we are waiting for third party testing. Still, from what we're seeing out of AMD, they're no longer down 10% in like-for-like comparisons. They're +/- 1% now. While being substantially cheaper. If the accompanying motherboards are competitive in features and build quality, Ryzen is a serious contender for all buyers, not just seriously budget-conscious buyers. It's no longer a matter of "oooh, I guess I can put up with not having the best to save some money." It's now "ooo, I can get exactly the same performance for half the price, and better peripheral support." (Well, I say now, but I mean March 2nd.)

      AMD fanboys can place pre-orders without even a hint of remorse or compromise. The rest of us can hold off on any planned new system purchases until mid-March, when the folks at Tom's Hardware are done with their benchmarking. Odds are that unless you really really really have to buy the Intel system because you don't intend to use it for anything other than running that ONE piece of software that is an outlier in benchmarks (whatever it might be: 7-zip?), then you should be buying an AMD system if you can find a motherboard that meets your needs. Unless you really enjoy throwing away $600 for nothing.

    6. Re:Yet more skewed bullshit testing from AMD. by JustNiz · · Score: 1

      >> I can get exactly the same performance for half the price, and better peripheral support."

      Please explain what you mean by better peripheral support?

    7. Re:Yet more skewed bullshit testing from AMD. by JustNiz · · Score: 1

      >> The only thing higher end is the 6950X

      Well, only in parallel tasks. The 6950x is only a 3GHz device. Single-threaded stuff will suck compared to anything modern. By comparison even the most ghetto $58 kaby lake desktop cpu is running at 3.0.Ghz.

    8. Re:Yet more skewed bullshit testing from AMD. by Anonymous Coward · · Score: 0

      Considering the numerous times that Intel has been caught cooking benchmarks, this is really funny. Intel's compiler is famous disabling optimizations if it doesn't detect a "genuine" Intel chip, even if the supporting features are still present.

    9. Re:Yet more skewed bullshit testing from AMD. by Anonymous Coward · · Score: 0

      I'd like to see a like-for-like benchmark between Ryzen and I7, such as single-thread at the same clock speeds.

      Uh, they did. The Cinebench single-threaded results are in the slide. Right hand side. The 1800X is indistinguishable from Intel's 6900K at single-threaded performance.

      Single thread test, boost clock is a realistic assumption here.
      - 1800X is running at 4.0GHz boost clock plus whatever XFR is able to add to it.
      - 6900K is running at 3.7GHz boost clock.
      Same result at 8% higher clock.
      Also, 6900K is Broadwell, mainstream 4-core Ryzen's will need to go against Skylake/Kabylake with higher clocks and slightly bumped IPC.

      Not saying that the result is bad, it's not. Frankly, it's awesome and shows AMD being clearly back in the race.
      But, from technical standpoint Ryzen is not ahead, it is slightly lagging behind (by a year or two) which is well compensated by lower prices.

      CAPTCHA: fooled

    10. Re:Yet more skewed bullshit testing from AMD. by Areyoukiddingme · · Score: 1

      Please explain what you mean by better peripheral support?

      24 PCI-e lanes between the CPU and the motherboard chipset, vs Intel's 16 in the same class, and therefore better able to saturate the throughput of modern high-bandwidth peripherals. Obviously not relevant for things like file transfers between NVMe and Gb ethernet because that's a DMA transfer where the data itself never touches the CPU, but helpful when the data requires actual processing prior to transfer.

  22. Finally! Will it last for AMD? by bobbied · · Score: 3, Insightful

    I think it's interesting that AMD finally got this CPU off the drawing board and actually onto silicon, finally. It has been a long time in development and has suffered many delays along the way, both from management changes and financial difficulty. They have put all their CPU eggs in this basket and I sure hope they have a good design here because Intel needs a bit of competition.

    I'm confident that AMD will make a go of this new architecture. It was a totally clean sheet design and has some unique and innovative features which may spur another round of slugging it out with Intel. What I find interesting here is the price point. Where I'm positive Intel has been racking in profit on their current offerings and will easily match AMD's prices, I'm hopeful that AMD will be able to press this new design into better performance than Intel can manage with their current technology at this price point.

    If history is any indicator, AMD will not be able to keep up once they wake the sleeping giant that's Intel. Where I'm not sure Intel really cares about the PC market (which is lagging a lot) they do care about profit. The question really becomes how much will this hurt Intel? I'm not sure it will be all that much, because Intel is diversified, doing lots of stuff in their own fabs. AMD has no fabs of their own anymore and really only have two business lines where they are the distant second player.

    Will it last for AMD? Will this put them back into an increasing market share and profitability? I hope so, but the guys over at Intel surely already have a good idea what they will do and what affect this will have on their bottom line. AMD may be off the mat, but they are seriously out classed by a company with deep pockets and technical ability.

    --
    "File to fit, pound to insert, paint to match" - Aircraft Maintenance 101
  23. LOL! From a coward who hates his own name by Anonymous Coward · · Score: 0

    See my subject: Don't attempt to patronize me loser - a man who hates his own real name's already dead you unidentifiable ac worm OR using a FAKE NAME ONLINE for your FAKE LIFE - you doing this trolling of me only PROVES I've utterly crushed you before many times.

    I'd say the "likes of YOU" should just commit suicide but you've already DIED a 1,000 deaths of a true coward (haven't you? Yes & ALL your wasted life, "ne'er-do-well")

    * Truer words were NEVER spoken on /. - you WISH you were me & could prove me wrong (but you never can or do).

    APK

    P.S.=> When ANY single "so-called 'competitor'" of mine can do more w/ less (as I do vs. them) NATIVELY? Then, talk (antivirus/dns/addons use more, do less & are FULL of security issues galore in the 1st 2 as well) - thanks for projecting "your kind" & INFERIOR inefficient limited competition of MINE can't handle ME, period, on ANY levels... apk

    1. Re: LOL! From a coward who hates his own name by Anonymous Coward · · Score: 0

      What makes host files better again? I don't think you mentioned that. Xd

    2. Re: LOL! From a coward who hates his own name by Anonymous Coward · · Score: 0

      His terrible hosts software requires a ton of resources to run apparently.

    3. Re: LOL! From a coward who hates his own name by Anonymous Coward · · Score: 0

      It's true. APK lack the programming skill to use anything other than Delphi's built in sort and believes it's impossible to do multithreaded programming with Delphi because somebody told him the runtime is not thread safe and he doesn't understand that only means you are responsible for the synchronization yourself. I've also read him trash a competing host file app for using SQLite saying that SQLite could just disappear at anytime (it won't, it's a widely used public domain software) while his app which takes hours to sort a few million list entries on a rather decent machine while Windows reports the app as not responding. Rather than correcting this issue he will stalk your user name on Slashdot and anywhere else he can find it to hurl childish insults and make grandiose proclamations about his skill. It's all very sad and points to some sort of untreated mental illness.

      Another possible explanation is that something more sinister lurks beneath the surface and his program is some sort of malware. That's why he's so obsessed with posting results from an online virus scan site but will not produce any further evidence other than unsubstantiated quotes supposedly hidden behind locked forums and quotes from Slashdot posters, many of which I've seen when they were written and are taken vastly out of context (usually a sarcastic quote repeated as an endorsement). I've have never seen his supposed quote from "Malware Bytes S. Burns, "I have seen the code and it is safe" anywhere other than in APK posts.

      Either way, there are many far better host files management applications out there, just search Google. Stephan Black's Host File Program seems pretty much equivalent in content but offers much better performance. This guy seems to be quite incompetent and uneducated when it comes to programming, you can tell by the language used in almost everyone of his posts, just a bunch of irrelevant bullshit. Even if his code isn't specifically malware I'd hate to trust code written by someone of such small intellect on my machine to begin with.

  24. I hope it is true. The CPU race by pjv936 · · Score: 1

    has been stagnant for quite a while. Hopefully this will restart it.

  25. Cheaper... For now. by TheFakeTimCook · · Score: 1

    While someone needs to put a stick in Intel's ass, I don't believe for a minute that this will remain a cheaper alternative, if AMD starts getting some traction.

    Oh, and I didn't see anything about power usage. AMD has always sucked in that regard.

    1. Re:Cheaper... For now. by Anonymous Coward · · Score: 0

      If the silicon is as described, it's the dominant option: better in every way. A lot cheaper, a little faster, a little lower power.

    2. Re: Cheaper... For now. by Anonymous Coward · · Score: 0

      I gave up using AMD CPUs about ten years so because they used so much power compared to Intel.

    3. Re: Cheaper... For now. by Anonymous Coward · · Score: 0

      So you ditched AMD for some sweet Netburst hot plate?

    4. Re:Cheaper... For now. by Arkham · · Score: 1

      While someone needs to put a stick in Intel's ass, I don't believe for a minute that this will remain a cheaper alternative, if AMD starts getting some traction.

      Oh, and I didn't see anything about power usage. AMD has always sucked in that regard.

      From the Ars Technica article:

      IPC is interesting in that it gives a sense of how cores are designed, but workloads aren't constrained by IPC or clock speed per se; they're constrained by thermal and power constraints. And AMD compares very favorably there, too: the Intel chip is a 140W part, so can use about 50 percent more power than the AMD.

      --
      - Vincit qui patitur.
    5. Re:Cheaper... For now. by TheFakeTimCook · · Score: 1

      While someone needs to put a stick in Intel's ass, I don't believe for a minute that this will remain a cheaper alternative, if AMD starts getting some traction.

      Oh, and I didn't see anything about power usage. AMD has always sucked in that regard.

      From the Ars Technica article:

      IPC is interesting in that it gives a sense of how cores are designed, but workloads aren't constrained by IPC or clock speed per se; they're constrained by thermal and power constraints. And AMD compares very favorably there, too: the Intel chip is a 140W part, so can use about 50 percent more power than the AMD.

      Hmmm, interesting. Thanks!

    6. Re:Cheaper... For now. by AvitarX · · Score: 1

      95 vs 145 watts is quite a little bit less power IMO

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  26. The one benchmark that matters by Anonymous Coward · · Score: 0

    Stop beating around the bush, and just tell me how long my Dwarf Fortress could go before it falls into the same ruin shared by so many others: FPS Death.

  27. Half price? Not for long by iTrawl · · Score: 1

    You think Intel is keeping the prices high because they can't cut them? Ha. They've arranged their line-up just so it looks like the i processors are good value for money. I imagine I'll see a bunch of Pentium processors withdrawn and i stuff prices sliced if AMD's CPU is that good and its price point is that low.

    Incidentally, that same price strategy is used by Apple. Keep an oldie in the line-up (Intel: Pentium; Apple: iPad mini 2) and then everything from that baseline to the top is price locked inside the range. Hurray, pre-owned stuff keeps its value for longer.

    --
    "Everybody's naked underneath" -- The Doctor
  28. Correct Name by Anonymous Coward · · Score: 0

    It's Dr. Lisa Su (https://en.wikipedia.org/wiki/Lisa_Su), not Lisu.

  29. Why so expensive? by Espectr0 · · Score: 0

    499 bucks? How will this be 20-50% lower than Intel? Intel i7-7700 is around $362.73 according to pcpartpicker, which is still lower than the lower spec $399 AMD.

    1. Re:Why so expensive? by Moof123 · · Score: 3, Insightful

      8 cores vs 4 cores. You gotta compare the equivalent number of cores, then kvetch over different clock rates and other details to really compare.

      Ryzen 1800X (8 cores) = $500
      Intel's consumer grade i7-6900K 8 cores (latest available) = $1000 (i.e. Ryzen is 50% lower than $1000)
      Intel® Xeon® Processor E7-4809 v4 (8 core) = $1600 (highway rape...)

      To compare 4 core vs 4 core:
      Ryzen 1400X (4 core with hyperthreading) = $200
      Intel 7700K (4 core with hyperthreading, though higher clock rate) = $340

    2. Re:Why so expensive? by Misagon · · Score: 2

      Intel has variants of the Core i7 7700 that have lower or the same clock as AMD Ryzen 1400X, but none of those costs less than $300 either.

      --
      "We mustn't be caught by surprise by our own advancing technology" -- Aldous Huxley
  30. Re:Finally! Will it last for AMD? by sl3xd · · Score: 1

    I'm hopeful that AMD will be able to press this new design into better performance than Intel can manage with their current technology at this price point.

    Your optimism is cute, but unwarranted. AMD won't really be a threat until they can easily beat Intel at every performance level, price be damned.

    Intel has no problem dropping prices to match the price/performance of anything AMD does. Inel is able to demand the prices they currently do because of AMD's decade-long streak of incompetence. Intel knows they can charge more because AMD isn't a threat.

    If you want to know what Intel does consider a threat, you have to look at the ARM architecture (and ecosystem). Granted, ARM isn't a single company, but that's part of Intel's problem: Competing with ARM is fighting a Hydra - one company's failure doesn't really affect the ecosystem much, and the combined might of all of the ARM licensees ensures its viability.

    ARM has almost perfect dominance for everything mobile and embedded - a market Intel has been trying desperately to enter, with zero success.

    --
    -- Sometimes you have to turn the lights off in order to see.
  31. Windows 7 & AMD CPU? by Anonymous Coward · · Score: 0

    If it runs Windows 7 (need that for gaming) then I'm in if not then I will stay for now on the i7...

  32. Compatability issues? by JustNiz · · Score: 0

    As a software developer where stability, reliability and conformance to x86 standards are critical, Intel has always been my CPU of choice. Also as a gamer seeking the most powerful system possible, Intel has always been my CPU of choice. The only criteria AMD have ever really won on is "Performance per $", which given that I am not particularly budget-limited has never been a significant consideration to me. Certainly not nearly enough to outweigh the others.

    Furthermore over decades of working with PCs I have seen a clear pattern that AMD-based PCs always seem to have little quirky things going on, such as more unexplainable blue screens etc, that just doesn't happen with Intel CPUs.

    In my experience, exactly the same can be said of their GPUs compared to Nvidia's too.

    Its nice to see AMD finally kicking Intel in the performance pants, because Intel are definitely overdue for some real competition at least in order to not just stagnate, however it seems to me that AMD even after decades still have other quirky little quality/reliability/compatibility issues to address that are just as damaging to their potential market deamnd as outright performance.

    1. Re:Compatability issues? by Jerry+Atrick · · Score: 1

      As a game developer and gamer, I remember AMD invented the X64 architecture Intel licences. Intel have been pretty good at conforming to AMDs standard. ...also looking at those AMD powered XB1 devkits on my desk.

    2. Re:Compatability issues? by JustNiz · · Score: 1

      >> AMD invented the X64 architecture

      My understanding is that AMD64 is really just an extension that adds 64 bit addressing mode and widened/added a few extra (64 bit) registers to the existing intel x86 spec.

    3. Re:Compatability issues? by barbariccow · · Score: 1

      It also codified a lot of extensions (like intel's SSE and SSE2) as part of the core.

    4. Re:Compatability issues? by Anonymous Coward · · Score: 0

      My desktop PC purchases have always primarily been driven by "Performance per $" which has led them all to have AMD cpus. I've never noticed them to have any more quirks than the Intel cpu-based computers I've used at work.

      Unfortunately, this may only hold true in the homebuilt market. Because when you buy prebuilt desktops, you often find the AMD cpu in the lowest-spec PC, which presumably has the cheapest possible motherboard and other components. This is an unfortunate reality, but not inherent to the AMD cpus.

  33. Is Ryzen pronounced "risen"? by Anonymous Coward · · Score: 0

    Is Ryzen pronounced "risen" (as in "AMD has risen from the dead")?

  34. doink shilligity speaks for us all by Anonymous Coward · · Score: 0

    >"You really have not needed a faster CPU since the launch of Sandy bridge."

    praise be, joe pc pedestrian has spoken for us all. L-O-L

    you v4pid cvnt, you obviously haven't spent hours waiting for some big app to compile, wanted to run dozens of VSTs in a DAW, or tried to do some 3D rendering. because if you had, you wouldn't make such ignorant lam3r declarations.

    you sux0r the corporate pen1s, you kneepadder trash

    it's pigs like you that make the world the open sewer it is today.

    can't you show me anything but surrender?
    obviously not, brownnosers are MT heads & a qt of saliva

    1. Re:doink shilligity speaks for us all by Anonymous Coward · · Score: 1

      Christ man are you having a seizure? Should we call a doctor?

    2. Re:doink shilligity speaks for us all by Anonymous Coward · · Score: 0

      I think way to late for that by the looks of it. He already equates his niche scenarios to what the general public need/want so he has some mental deficiencies. It is not an uncommon delusion sadly, he thinks big companies should be desperate for his $1 instead of the $1 billion dollars they get for addressing the needs of the majority. There will always be some scenario that requires bigger, better, faster but unless it makes commercial sense arrogant arseholes like him can just do without.

  35. I'll let /.ers speak for me #1/2 by Anonymous Coward · · Score: 0

    I'm going to continue using the Host File Engine. Your software is well written, functional. The Host File Engine performs exactly as promised by mmell

    his hosts program is actually pretty good by xenotransplant

    his hosts tool is actually useful for those cases in which one does indeed want to locally block stuff outright while consuming minimum system resources by alexgieg

    I've never tried to belittle (APK's) work, I've flat out said it's good by BronsCon

    take a look at the APK hosts file engine by SuperKendall

    APK is kinda right. I've tried his hosts file generating software. It works by bmo

    APK is totally right on this count. Adblock Plus on Firefox mobile is a dog on older, or lower end, phones. A hostfile based adblocker makes for a much better experience by chihowa

    I like your host file system by Karmashock

    I find your hosts file admirable by vel-ex-tech

    * My code's liked + recommended & hosted by Malwarebytes' hpHosts!

    APK

    P.S.=> More coming... apk

  36. I'll let /.ers speak for me #2/2 by Anonymous Coward · · Score: 0

    I support APK's stand on the hosts file by Trax3001BBS

    Your premise that hostfiles are a good way to deal with advertising and malvertising is quite valid by JazzLad

    No complaints from me, I like APK... Reminds me to use a host file. Also, his stuff is free by aaaaaaargh!

    APK's monolithic hosts file is looking pretty good by Culture20

    APK... Awesome to see he's still spreading the good word by Molochi

    ABP is insufficient as a solid hosts file does everything that APK reminds us about by fast turtle

    APK isn't wrong by cfalcon

    APK, I know people give you a lot of shit regarding hosts, but please don't ever stop by nasredin

    You need APK's hosts file by Teun

    APK solution STILL relevant by Thud457

    you're right about hosts files by drinkypoo

    APK

    P.S.=> They're in addition to https://hardware.slashdot.org/comments.pl?sid=10280229&cid=53912657 many more earlier + 1,000's worldwide - there's no arguing w/ success... apk

  37. I'll let /.ers speak for me #1/2 by Anonymous Coward · · Score: 0

    I'm going to continue using the Host File Engine. Your software is well written, functional. The Host File Engine performs exactly as promised by mmell

    his hosts program is actually pretty good by xenotransplant

    his hosts tool is actually useful for those cases in which one does indeed want to locally block stuff outright while consuming minimum system resources by alexgieg

    I've never tried to belittle (APK's) work, I've flat out said it's good by BronsCon

    take a look at the APK hosts file engine by SuperKendall

    APK is kinda right. I've tried his hosts file generating software. It works by bmo

    APK is totally right on this count. Adblock Plus on Firefox mobile is a dog on older, or lower end, phones. A hostfile based adblocker makes for a much better experience by chihowa

    I like your host file system by Karmashock

    I find your hosts file admirable by vel-ex-tech

    * My code's liked + recommended & hosted by Malwarebytes' hpHosts.

    APK

    P.S.=> More coming... apk

  38. Power Consumption by Anonymous Coward · · Score: 0

    Do these work with powerd?

    What does power consumption on an idle box look like?

    1. Re:Power Consumption by Anonymous Coward · · Score: 0

      > What does power consumption on an idle box look like?

      Why would you let a perfectly good CPU sit around idle?!
      Let it do something at 100% before its obsolescence, because that's what it was designed and manufactured for!

  39. I'll let /.ers speak for me #2/2 by Anonymous Coward · · Score: 0

    I support APK's stand on the hosts file by Trax3001BBS

    Your premise that hostfiles are a good way to deal with advertising and malvertising is quite valid by JazzLad

    No complaints from me, I like APK... Reminds me to use a host file. Also, his stuff is free by aaaaaaargh!

    APK's monolithic hosts file is looking pretty good by Culture20

    APK... Awesome to see he's still spreading the good word by Molochi

    ABP is insufficient as a solid hosts file does everything that APK reminds us about by fast turtle

    APK isn't wrong by cfalcon

    APK, I know people give you a lot of shit regarding hosts, but please don't ever stop by nasredin

    You need APK's hosts file by Teun

    APK solution STILL relevant by Thud457

    you're right about hosts files by drinkypoo

    APK

    P.S.=> They're in addition to https://hardware.slashdot.org/comments.pl?sid=10280229&cid=53913023/ many more earlier + 1,000's worldwide - there's no arguing w/ success... apk

  40. AMD 95 Watts VS Intel 140 Watts by BrendaEM · · Score: 1

    If I were intel, power consumption would scare me. Intel has no competing part power-wise. It all adds up.

    --
    https://www.youtube.com/c/BrendaEM
  41. I for one by Anonymous Coward · · Score: 0

    look forward to our piledriver- and ddr3- selling ebay overlords.

    Time for some cheap upgrades.

  42. received pre-order pricing from Micro Center by Anonymous Coward · · Score: 0

    FYI, this is not an endorsement. But here is what a Micro Center email today quoted for pre-order pricing:

    Ryzen 7 1800x, $499.99, sku 251736
    Ryzen 7 1700x, $399.99, sku 250837
    Ryzen 7 1700 , $329.99, sku 250803

    I have no idea how these prices compare to other vendors.

  43. Yawn by Anonymous Coward · · Score: 0

    When will 3D CPU's finally be a thang?

    ps- yay for amd

  44. Not only are the CPUs less expensive... by MoarSauce123 · · Score: 4, Informative

    ....AMD mainboards are noticeably less expensive. Intel is the synonym for overpriced.

  45. You've done better "Flimzor" (lol)? No by Anonymous Coward · · Score: 0

    Many /.ers like/use my work quoted (malwarebytes' hpHosts hosts & recommends it + verified it safe @ code level) https://hardware.slashdot.org/comments.pl?sid=10280229&cid=53913023/ & https://hardware.slashdot.org/comments.pl?sid=10280229&cid=53913235/

    My code's safe by 57++ antiviruses https://www.virustotal.com/en/file/e01211ca36aa02e923f20adee0a3c4f5d5187dc65bdf1c997b3da3c2b0745425/analysis/1433430542/

    * A jealous "FLIMZOR" ("ne'er-do-well" nobodies doing unidentifiable ac troll posts like you) doesn't stand out @ all + has accomplished zero (& you know it).

    APK

    P.S.=> Hostsman can't do hardcoded favs resolution (faster vs. LAN or remote DNS) where people spend most time online & isn't 64-bit. Mine is & nothing does hosts better vs. APK Hosts File Engine 9.0++ SR-7 32/64-bit https://www.google.com/search?hl=en&source=hp&biw=&bih=&q=%22APK+Hosts+File+Engine%22+and+%22start64%22&btnG=Google+Search&gbv=1/ ... apk

  46. Re:Finally! Will it last for AMD? by Anonymous Coward · · Score: 0

    The actual threat is the fabless business model. Qualcomm and Apple are two companies proving that it is very profitable if your IP is good enough. IBM divested their fabs because they recognised the spiralling R&D costs are much harder and riskier to recoup when you only have your own products to sell.

  47. A little soon by dfeifer · · Score: 1

    Pre orders available starting today but the official launch isn't until march 2nd.

  48. Checked out Mr. Black's imitation by Anonymous Coward · · Score: 0

    Inferior IMITATION = sincerest form of flattery written in python interpreted non-gui code (python doesn't do that by itself - it's a scripting language @ best/most for noobz & Yes I code in it myself sometimes & can offer that estimation of it - limited)

    AND above ALL else?

    Yes, HIS CAME OUT MUCH LATER THAN MINE DID proving it's MERE imitation, nothing more!

    (... & using python, he didn't write anything himself really - the language itself provides the commands he merely used (& as far as sorts? I used the MOST optimal sort there is quite possibly in default quicksort (though I have written code for others that are better on small/mostly sorted OR unsorted, large/mostly sorted OR unsorted - uncessary & NOT optimal (heapsort's nice though, but it's HUGE code I don't need)).

    APK

    P.S.=> Above ALL else though? I'd give him credit for doing SOME kind of work, but you?? LOL, no way 'flimzor' the unidentifiable ac ZERO that you are (that I've obviously CRUSHED before head to head & so you do these ac posts of yours - how can a SCUMBAG whimp like YOU live with yourself I wonder?)... apk

  49. 10 core is the new way to go by Anonymous Coward · · Score: 0

    8 core is old news.

    What about Intel's 10 core CPUs?

  50. Re: frist ps0t by Anonymous Coward · · Score: 0

    *Yawn*

  51. Windows 10 by Anonymous Coward · · Score: 0

    I would be more enthusiastic except for the fact that they've said Ryzen will only work on Windows 10.

    Maybe we'll see unofficial drivers for Windows 7 or maybe they didn't pull an Intel and made the hardware backward compatible with Win7-level drivers, but given that I'm boycotting Windows 10, I have no compelling reason to upgrade as my existing APU systems are not even taxed by the stuff I use under Linux.

    The only real drive for me to upgrade is games performance and that is still a Windows world...