AMD Set To Launch Ryzen Before March 3rd (anandtech.com)

An anonymous reader shares an AnandTech report: Thanks to some sleuthing from various readers, AMD has accidentally let the cat out of the bag with regards to the official Ryzen launch date. While they haven't specifically given an exact date, the talk to be given by AMD at the annual Game Developer Conference (GDC) says the following: "Join AMD Game Engineering team members for an introduction to the recently-launched AMD Ryzen CPU followed by advanced optimization topics." The GDC event runs from February 27th to March 3rd, and currently the AMD talk is not on the exact schedule yet, so it could appear any day during the event (so be wary if anyone says Feb 27th). At this time AMD has not disclosed an exact date either, but it would be an interesting time to announce the new set of Ryzen CPUs right in the middle of both GDC and Mobile World Congress which is also during that week. It would mean that Ryzen news may end up being buried under other GDC and smartphone announcements.

Comments

AMD has

[sirber]

Ryzen from the dead!

Re:AMD has

[Anonymous Coward]

I hope AMD succeeds.

We need competition on the x86 CPU space. Intel dominates the top end of the x86 space right now, and they have no incentive to come up with better CPUs. Today's top of the line CPU is just last year's top of the line CPU with a few features tacked on and less power consumption (which sounds great on paper, but in practice takes roughly 10 years for the savings to materialize... and by then, you've already swapped your CPU 3 times).

But I don't think AMD can pull it off. It's been about 10 years since I've been excited about one of their products and they've been over-promising and under-delivering for a long time.

But we'll see.

Re:AMD has

[Anonymous Coward]

We need competition on the x86 CPU space.

why? the x86 market is contracting. how does adding competitors work out in a contracting market? so you like blood in the water?

Re:AMD has

[barc0001]

> in practice takes roughly 10 years for the savings to materialize... and by then, you've already swapped your CPU 3 times

Actually these days, not so much. I've got an i5 2500K that I bought in early 2011 in my home workstation, and I have no plans to replace it any time soon. My general rule is that I won't replace a processor unless it's both old and I will get around twice the performance of the old one. Looking at what I'd replace it with if I was to build the same machine today - an i5 6600K - there's just no point. I'd get about a 50% boost over what I have, and what I have is already more processor than I need for just about anything I do with the exception of gaming. And for that the money is better spent on a new video card, and that's what I do replace every 2-3 years.

In the past with Moore's Law that was around every 2 years, but Intel's been stagnant on progress for so long, they're now running ads like this:

http://www.ncix.com/article/intel_kabylake.htm

Oooo... up to 28% better performance than a 3 year old part! And all you need to do is replace your chip, motherboard and probably RAM. Pass. Instead of spending $600 on all that I'll just drop $200 on last year's hot high end video card.

Re:AMD has

[Anonymous Coward]

As you can see below, AMD needs to add support for ECC memory.
Intel's skylake xeon e3-1275v5 and skylake i3's are no joke.
And everyone overlooks them.
All of my builds now use the e3-1275v5 or i3-6320, those are the real sweet spots.
And they all use ECC memory... same performance, ram cost is the same by avoiding artificially high[end] i7 cpu prices, and no more data corruption :)
Plus hyperthreading, vt-d, aes, etc... they're a total win.
If AMD zen/ryzen has ECC, it will squarely compete against these two cpus.
If not, AMD loses.

http://ark.intel.com/Search/Advanced?s=t&RetailSkuAvailable=true&FilterCurrentProducts=true&ECCMemory=true&VTD=true&AESTech=true
http://ark.intel.com/Search/Advanced?s=t&RetailSkuAvailable=true&FilterCurrentProducts=true&ECCMemory=true&VTD=true&AESTech=true&CodeNameText=skylake

Re:AMD has

[TsuruchiBrian]

The fact that Intel is simply cashing in on their past success is probably the thing that will allow AMD to catch up. I don't think Intel even wants to blow AMD out of the water. This would cost more money and bring on monopoly accusations. That said, I'm glad we seem to be in the part of the cycle where AMD might overtake Intel and spur some competition. The other part of the cycle is boring.

Re:AMD has

[Anonymous Coward]

who da fuck cares about desktops... soon even programmers will only use a phablet and a bluethooth keyboard. Normal people only use phones now anyways. Heck in the last months gfx designers I have seen were using ipads. Garageband is available on iphones. You just need a bluetooth sound projector and you can be the next David Guetta. Once ARM wins the race they will just develop a server CPU family and destroy both intel and amd...
Simply, intel and amd is the past.

Re:AMD has

[Anonymous Coward]

I'm still using a 2600k. I've upgraded my video card a couple times since I built this machine, replaced the motherboard, and gotten a couple more hard drives and an SSD. CPU upgrades just haven't been worth it.

However, the features on this motherboard are outdated af. I will be upgrading this year just because I'm so far behind on features. The 50-60% performance boost will just be a nice bonus.

Re:AMD has

[unixisc]

The x64 is no longer the only game in town. Aside from ARM, which has been gaining traction, there is also RISC V coming along, and so there will be alternatives in the market. What's more - since Android seems to be in the driver's seat, rather than Windows, x64 has even less clout going forward.

The fact that more semiconductor houses can make ARMs or RISC Vs, but only Intel and AMD the x64 means that those CPUs would be more attractive to vendors to support than x64. Don't be surprised to see even Microsoft migrate to these alternate platforms

Re:AMD has

[Ayanami_R]

No incentive or are physics in the way?

Serious question because I am doubtful that they're just "not advancing" just because of "no competition" But rather I don't think it's possible to move as fast with performance increases now because we're rapidly approaching the limits of current silicon designs can do.

Re:AMD has

[Anonymous Coward]

There is already ARM CPUs for servers.

Re:AMD has

[Anonymous Coward]

If you're having routine data corruption with non-ECC RAM then you're doing it wrong.

Re:AMD has

[epyT-R]

A narrow vision... You will become one with the Borg. You will all become one with the Borg.

Re:AMD has

[m.dillon]

Unfortunately, Google did a study that showed that routine corruption without ECC is a fact of life. It is much more prevalient than people thought.

-Matt

Re:AMD has

[Anonymous Coward]

Intel dominates the x86 space

FTFY.

fact is, intel has dominated all segments of x86/x64 since the first-generation 'core' architecture. before that, scrappy amd with its superior dual core athlon (intel's glued-together dual core netburst chips just couldn't compete) was clawing away at a near 50/50 market share split with intel. but it's been intel only ever since, and all downhill for amd.

dollars, watts, or outright performance. you choose the metric, it does not matter. intel owns the ENTIRE pc processor market, period. do not let anyone tell you otherwise.

the only people that buy amd processors are either clueless suckers buying on price alone, fanboys, or anti-intel for whatever reason.

if this new architecture from amd is just another disappointing overhyped suckfest, like everything else they've come up with the last 10 years, they are D-O-N-E *done* with the pc market.

Re:AMD has

[Anonymous Coward]

You are correct, however, I am not sure you understand how many BILLIONS of $$$ have been invested in x86x/64 infrastructure.

For most large multinational corporations, they can`t just flip the switch and change to ARM platforms....At least not until ARM is capable of running x86 Linux/Windows VMs.

Also I like to think that Intel and AMD are not that stupid. They know that right now, ARM beats out their platforms on power consumption which is why they are kings in mobile.

If ARM comes close to matching their performance on servers/workstations, IMHO you will likely see Intel/AMD either dropping their prices or perhaps switching to licensing x86-64 designs to other manufacturers .

Also there is the issue of interoperability between ARM processors. Will the same OS and apps run on an ARM CPU made by Samsung vs Texas Instruments? I have no idea, hypothetically it should but realistically that is not the way it works right now (ever tried running a firmware intended for a different phone on your android).

Re:AMD has

[Anonymous Coward]

Unfortunately, Google did a study that showed that routine corruption without ECC is a fact of life. It is much more prevalient than people thought.

...and yet it has no quantifiable effect on real life.

Re:AMD has

[The-Ixian]

dollars, watts, or outright performance. you choose the metric, it does not matter. intel owns the ENTIRE pc processor

Ok. I choose dollars.

the only people that buy amd processors are either clueless suckers buying on price alone, fanboys, or anti-intel for whatever reason.

Did I win?

Seriously, though. The only metric I need is: Does it perform well enough for my modest gaming/computing needs?

My AMD processor from 2014 (or thereabouts) still does the job admirably. I really don't care if it is inferior to an Intel chip.

If I can get the computing power I need for a cheaper price, why in the world would I choose to buy something more expensive?

Re:AMD has

[Anonymous Coward]

Ryzen news may end up being buried under other GDC and smartphone announcements.

Also, the news shall ryze from the under! They wouldn't be ryzen news otherwise.

Re:AMD has

[jwhyche]

Lets just stop and be serous here for a bit. Desktop sales may have slown down but desktops are not going any where. Nobody programs on tablets unless they are developing apps for them. Even then most of the development takes place on desktops in a emulator. I count a docked laptop as a temporary desktop.

Real gamers do no use phones for games. Real gamers use custom made desktops with high end graphics cards and big fat monitors. They have surround sound audio and carefully selected controllers.

ARM, while a good processor design, isn't really found outside the phone and tablet market. You find some low powered chrome books that use it but that is about it.

Nobody in the industry takes a ARM server seriously. You might have a few "mavericks" that use it but the real server market is dominated by the x86 servers. An it will continue to be for the foreseeable future.

Re:AMD has

[Khyber]

AMD processors have supported ECC for a long time (starting back in the AM2 days.) It's on the motherboard maker to support it, not AMD.

Re:AMD has

[KingMotley]

I think his point is that the $x intel CPU is faster, and uses less power than the equivalent $x CPU from AMD for every value of x. Why did you by AMD?

Re:AMD has

[Anonymous Coward]

Unless you're running it 24/7 at 100% for a few years the extra power consumption isn't going to make a difference.

Re:AMD has

[unixisc]

I know what's been invested into the x86 infrastructure: however, the point I was contesting was that the existence of dual sources for the x86 is essential for the market

As Apple illustrated some 10 years ago, they could switch from PowerPC to the x86 pretty smoothly, when they weren't happy w/ the architectural roadmap of the former. The same is true for x86 vs ARM, or even other CPUs. Simple reason for that: while in Apple's case, they had dual ported NEXTSTEP to not just PowerPC but x86 as well while creating OS X, the same is true for most of the OSs in question, and the only exception is Windows, which has frankly been struggling the last couple of years.

Note that for things like ARM, they already have Linux and BSD, which are the 2 main widely ported platforms, which makes a migration in case of a weakening of the x86 infrastructure not as bad as it used to be in the 90s. During that time, one would have had to migrate from not just one CPU to another, but also one OS to another e.g. Intergraph CLIX/Clipper to Sun Solaris/SPARC. Which was typically a pretty major undertaking. In sharp contrast, not only does one have a common OS, w/ open sources to handle any wierdness in the CPU instruction sets themselves, but there are also things like bytecode and virtual machines that didn't exist 20 years ago. Making the migration process a lot smoother.

The only case where it can be argued that portability would be a problem would be Windows, but even there, they've had Windows RT, Windows 10 Mobile and Windows 10 IoT. So it can be argued that all 4 major platforms - Windows, Linux, Mac OS and BSD - are there on both x86 and ARM, and if Intel tries to horn in on AMD's decline, they'd get slammed by the likes of Qualcomm, NVIDIA, NXP and everyone else. Which is why x86 is not the priceless gem that it was even 10 years ago, when AMD sabotaged Intel's Itanium plans by extending the x86 instruction set to 64- bit

Also, Intel's biggest competitor is not even AMD, nor is it ARM. Over the last several years, their biggest competitor has been their own installed base. They have made CPUs that are so good that there ain't a compelling reason to upgrade, and it's that which has caused a relative period of stagnation for them. So they need to explore which new markets to enter and using which vehicles, and they need to make that vehicle competitive enough to take care of both their top and bottom lines

Re:AMD has

[Anonymous Coward]

I have used AMD processors for years just for the cost/performance/ECC support. With Intel the issue has been the cost and particularly the availability of workstation motherboards in our local markets.

Re:AMD has

[Anonymous Coward]

At least not until ARM is capable of running x86 Linux/Windows VMs.

Why would you bother virtualizing linux just to run on x86?

you will likely see Intel/AMD either dropping their prices or perhaps switching to licensing x86-64 designs to other manufacturers .

Not going to happen. You would instead just see them build their own ARM processors, which both have done in the past, most recently AMD.

Also there is the issue of interoperability between ARM processors. Will the same OS and apps run on an ARM CPU made by Samsung vs Texas Instruments? I have no idea

Clearly you have no idea, because that is not an issue, unless you're talking major architectures, like v7 or v8, and even then there is reverse compatibility.

Re:AMD has

[Anonymous Coward]

No incentive or are physics in the way?

Serious question because I am doubtful that they're just "not advancing" just because of "no competition" But rather I don't think it's possible to move as fast with performance increases now because we're rapidly approaching the limits of current silicon designs can do.

We're not done advancing, it's just that Intel is running into issues it hasn't really bothered addressing, likely due to the cost. I am quite certain that the lack of competition has led Intel to keep going with the same basic core design for much too long.

If AMD is successful with their new architecture, you'll undoubtedly see Intel have to come up with a new architecture as well.

Re: AMD has

[tgrigsby]

My main box uses an AMD Phenom II X6 3.7GHz. Built in 2009, I've upgraded the memory, graphics, and RAID array, installed a USB 3.0 card, and replaced a burned out DVD drive, but the CPU and motherboard still kick ass.

Oh, and it runs Windows 7 Ultimate. I win.

Re:AMD has

[strstr]

What the poster is saying is there is no reason not to have ECC. Corruption of bits is very easy and its one of the secret NSA hacking tools (space based weapons corrupt- read- write RAM).

When all you need is a few extra bits and some correction mechanism why the fuck not use it? It costs the same to make too.

ECC unfortunately cannot stop space based hacking but it can guard from some attacks at least.

https://www.obamasweapon.com/

Re:AMD has

[Anonymous Coward]

Can you add some PCIe 1x or 4x cards to get some features you want/need?

Re:AMD has

[Anonymous Coward]

I think it was slashdot comments that told me what ECC was really useful for. When an error happens, it will get logged. So if you ever have an error you might know about it, and if it happen way more often than say, once a year, then you can replace a stick before something bad happens I guess (or underclock your RAM, I used to do that trick to run my PC with "dead" RAM).
So, hardware semi-failure would be the scenario, it would be more useful/common than protection against cosmic rays, which is a particular story I don't give a damn about (some bloke said years ago the RAM cells got so tiny, a cosmic ray will hardly ever hit whatever thing it has to hit to mess it up)

If you deal with encrypted files or compressed files or file systems, maybe it's easier to get some of your data ruined, I wonder.

Re:AMD has

[Anonymous Coward]

And nobody cares much, because it turns out increasing the server count is not always a winning proposition, and the lower single-thread performance increases the latency of individual requests/operations, which likely run some horrible code written by mediocre developers.

Re:AMD has

[Anonymous Coward]

I've got an i5 2500K [...] what I have is already more processor than I need for just about anything I do with the exception of gaming.

TBH, you just need to OC it to over 4GHz, and if you can hit close to 5GHz, it'll beat many modern CPUs. I'm using an OCed i5 2500k in my 4k gaming rig without issue.

Just buy a decent heatsink or AIO (all in one) watercooler, then go into your BIOS to increase the voltage and multiplier. Do some stress tests to ensure it's stable, and you're done. Some motherboards offer automatic CPU OCing too; look for "stages" or something like that in the BIOS.

I bought a used AIO Corsair watercooler and threw on 2 Gentle Typhoon 120mm fans (push/pull config). I have it OCed at 4.8GHz and only one core ever goes over 50C under load (live in warm Southern California). I run this frequency 24/7.

Using a ASRock mobo (Z68 Extreme7 Gen3) and my old ass PC Power and Cooling 610W PSU.

In games, CPU utilization is laughable.

Re:AMD has

[Agripa]

AMD processors have supported ECC for a long time (starting back in the AM2 days.) It's on the motherboard maker to support it, not AMD.

AMD's dedicated CPUs support ECC on the various AM sockets. Their APUs unfortunately do not and it is not clear if Ryzen does.

Re:AMD has

[Agripa]

I have used AMD processors for years just for the cost/performance/ECC support. With Intel the issue has been the cost and particularly the availability of workstation motherboards in our local markets.

I had the same reason to switch to AMD. When I built my last workstation to replace my Pentium 4 ECC system, the Intel premium for ECC was more than $1000 because all of the Xeon based systems used FB-DIMMs.

Intel ECC systems still cost more than AMD ones but the premium is down to $100s of dollars.

Re:AMD has

[Agripa]

There is already ARM CPUs for servers.

There sure are. And the one available ARM workstation board cost 3 times as much as an Intel workstation board yet has pitiful I/O, memory, and performance ... sort of like an Apple laptop.

Intel needs a swift kick in the ass

[Master5000]

But history shows AMD hasn't been able to be that kick in the past 10 years.... Wait and see is my recommendation. Don't go by the hype!

Re:Intel needs a swift kick in the ass

[slashdice]

also, don't go buy the hype.

Re:Intel needs a swift kick in the ass

[KiloByte]

But history shows AMD hasn't been able to be that kick in the past 10 years....

5 years ago I did the research and it was AMD Phenom II x6 which was at the sweet bang-to-buck spot. Worse in single-threaded but for compiles/etc kicked the ass of similarly priced Intel chips.

And despite its age I still don't feel an urge to upgrade -- it's adequate for software I work with, I don't compile anything bigger than the kernel these days. So unless I'll need to run big bisects or something, it stays for now. Being one of the last non-backdoored CPUs from both sides of the aisle is a big bonus.

Re:Intel needs a swift kick in the ass

[Anonymous Coward]

so it could appear any day during the event (so be wary if anyone says Feb 27th)

My bet is February 29th

Re:Intel needs a swift kick in the ass

[twistedcubic]

My AMD Athlon II x4 is from 2010, and I still use on the Desktop (24GB ECC RAM, recent mobo upgrade only). I also don't feel the need to upgrade, but the hardware encryption from all the modern processors would be nice to have.

Re:Intel needs a swift kick in the ass

[Kjella]

also, don't go buy the hype.

Well, for the most part you can't actually buy the hype it's all pre-release, post-release benchmarks and reviews dominate. I'm not sure what the point of it is really, maybe it'll make a few people hang on to their old gear a tiny bit longer ahead of the release but I doubt it. I mean Apple newer showed previews of future products and I'd call them moderately successful. I guess it's more to please investors and creditors than customers, things are not so hot right now but right around the corner is the pot of gold. The problem is you often end up with a light Osborne effect, most people don't want the product you have now on the shelf they want the next one. And when that's for sale, they want the next one after that. That's not very good if you want to actually make sales.

Re:Intel needs a swift kick in the ass

[TsuruchiBrian]

I'm holding off on buying a new chip until ryzen comes out. Maybe it wouldn't make sense for Intel to do this kind of thing, since they might even prefer you buy one of their existing chips rather than waiting for the new one to clear out some old inventory, but for AMD, you might be holding off on buying a current Intel product to wait to buy a new AMD product. I'm sure as hell not buying an excavator now, and I don't think anyone else is either, and I think AMD knows that.

amd just needs to have more pci-e lanes with

[Joe_Dragon]

amd just needs to have more pci-e lanes then Intel without having 10-15 lanes + sata + network + pci-e storage all stacked off of the X4 dmi link.

If it gets bad reviews...

[Anonymous Coward]

... will they rename it "Fallyn"?

BOTH AMD & Intel need this kick in the ass

[Anonymous Coward]

See subject: This invalidates/neutralizes an entire class of exploit via a mirrored readonly callstack via "CET" https://www.google.com/search?hl=en&source=hp&biw=&bih=&q=Intel+CET&btnG=Google+Search&gbv=1/

* THAT is well-worth doing + worth waiting for... blowing away buffer overflows & rip-rop attacks cpu side!

(Funniest part is, it IS already "doable" but they don't implement it for consumers (or anyone afaik @ least)).

APK

P.S.=> They do that? I'll buy again (even though I own a fairly new & powerful CPU in the Intel 4790k)... apk

Re:BOTH AMD & Intel need this kick in the ass

[NotAPK]

Hey APK, why are you going around misusing my username? I've worked hard on Slashdot to build my reputation and you come along and spout garbage and start flame-wars for no reason at all. Just do a brother a favour, back off and leave me alone. Deal?

Still no competition

[JuanDavila]

Where is the competition though? I still feel like as far as speeds go Intel wins. So what is keeping AMD alive, price and a new flashy name...?

Re:Still no competition

[Anonymous Coward]

I would say the fact that their processors compete with Intels $1,000 processors and will be expected to be between $500 to $800 for that level of performance would cause a great deal of competition.

They don't have to beat them and their highest of high end offerings, the just have to match them at the levels the consumers buy which they appear to be doing just that. So I expect to see some great competition between them with Intel also being forced to lower their prices to remain competitive with them.

Re:Still no competition

[drinkypoo]

Where is the competition though? I still feel like as far as speeds go Intel wins. So what is keeping AMD alive, price and a new flashy name...?

Price is enough. Half the performance for 25% of the price is a pretty compelling package for some of us, which is roughly how the current chips come out, maybe better. If the new chips are in the same ballpark, they'll sell like hot dogs.

Re:Still no competition

[Cmdln Daco]

As long as you're willing to settle for half the performance and also being perceived as really cheap.

Re:Still no competition

[Anonymous Coward]

Perfect answer which shows exactly why Intel fanboys can't be argued with.

AMD is bang for buck, Intel is not = Intel wins. People who buy Intel don't buy it because they need them, they buy Intel because they participate in a moronic dickwaving contest with themselves. They are just like the customers who buy thisapp for Android. "OMG can't be percived as really cheap". LOL

It's even so bad they have to spread lies to justify it to themselves, "half the performance". ROFL

Re:Still no competition

[TsuruchiBrian]

Ferraris are faster than Toyotas. That doesn't imply that Ferrari is doing better than Toyota as a company. Intel is doing better than AMD, but it's not because it makes the fastest chips. Very few people are buying the $1000 Intel chips. Intel is winning the $200-$300 chip market. AMD could very conceivably start competing or even start winning in this market even if it doesn't take the top spot.

Re:Still no competition

[TsuruchiBrian]

I don't think many people buy CPUs as a status symbol. So yes, I suspect many people will be ok with the stigma of owning a cheap CPU if it means they were actually able to spend less money on it, and it does what they need.

Re:Still no competition

[Anonymous Coward]

AMDs APUs is the best choice, and best bang for the buck, for 4 out of 5 desktop users. Only power users need either a fast, discrete graphics card, or a very fast CPU. Intel's offerings in the same segment have somewhat higher CPU performance, much weaker GPU performance, and at a much higher price - AMD clearly wins the APU segment.

Re:Still no competition

[m.dillon]

Its not 25% of the price though. Not sure where you got that from. The benchmark Intel cpu that AMD is competing against is the I7-7700K, which is $350 on Amazon. It will be the AMD 6-core against that one.

AMD will also be able to compete against Intel's i3's. An unlocked Ryzen *anything* (say, the 4-core ryzen) will be the hands-down winner against any Intel i3 chip on the low-end. Intel will have to either unlock the multipliers on all of its chips to compete, or pump up what they offer in their i3-* series.

The AMD 8-core is probably not going to be a competitive consumer chip.

-Matt

Re:Still no competition

[Anonymous Coward]

"Stigma"? You've got funny ideas son. If you overpaid for a CPU when a cheaper one could have done the same job, the only "status" that gives you is the "moron-flag". Only retards think it's cool to overpay. It's like chopping your dick off because it's supposedly the cool thing to do...

Re:Still no competition

[drinkypoo]

As long as you're willing to settle for half the performance and also being perceived as really cheap.

My current primary transport is a 1982 300SD, what do you think?

Re:Still no competition

[Cmdln Daco]

Pease drove a '67 bug and it ultimately killed him. Be careful.

Re:Still no competition

[drinkypoo]

Pease drove a '67 bug and it ultimately killed him. Be careful.

I'm working on upgrading to a 1998 A8 Quattro. It's the only model they bothered crash testing, but it got five stars so that's cool for me if not for later A8 buyers. The W126 300SD is pretty much the safest car of its era as well.

Re:Still no competition

[Anonymous Coward]

The Sandy Bridge, Haswell, Skylake Pentiums have been some heavy competition to AMD's low end, and the i3 6100 is a small beast with good enough graphics of its own. Pentium G4560, newly launched, is another problem for AMD, it's got hyperthreading and it's cheap.
So it's been quite rough.

Re:Still no competition

[Anonymous Coward]

I still feel like you're retarded, as far as IQ goes, I win. So what's keeping you alive, sucking dick and a retarded name?

It's coming Feb. 29th

[cheese_boy]

I have it on good authority that it'll be Feb. 29th.

Re:It's coming Feb. 29th

[Hydrian]

I'm thinking April 1st is more realistic.

Also the fact...

[Anonymous Coward]

That the older processors probably do a better job of 'respecting your rights', whereas the new ones from Intel, AMD, and ARM licensees all have Clipper Chip on steroid built in and very difficult if not impossible for the end user/hardware owner to disable.

Hackaday just posted an article that claims to be able to disable the Intel ME via firmware hacking, but in order to do it you need either a socketed bios chip, or a programmer with clip leads to read, modify and reprogram a soldered down one.

What that leaves is Broadwell on the Intel side (GPU firmware signing on either Broadwell or Skylake, plus the ME code.) and FM2 (not +) and AM3+ as the max on the AMD side.

Anything newer by either company, or anything touting trustzone that doesn't have an unsigned stage0/1 bootloader is essentially a trojan horse awaiting activation.

There is no sane reason for individuals or companies to be buying processors of newer generations that they cannot guarantee are secure, and yet people are continuing to do so with the weak and extremely dangerous excuse of 'Don't you think if they were going to do that, they would have done it by now?' rather than noting that giving a few generations for complacency before bringing down the hammer always works better to trap people inside technological or institutional oppression, by which point they no longer have a mechanism to get back out.

If you can, buy last generation systems that didn't have it, can be disabled, or can be circumvented while you still can, and hope I and others like me just get proven as unfounded kooks, rather than having our concerns brought to light in the sort of oppressive sweep we might not get another opportunity to step out of.

Re:Also the fact...

[Anonymous Coward]

Every AMD TrustZone enabled system I've ever encountered included an system BIOS setting to disable the feature. As long as you don't need Bitlocker encryption you have the option to turn it off if you don't want/trust it. Don't let that stop you from spreading your FUD though.

late

[edxwelch]

Wasn't it ment to have been launched in 2016?

Re:late

[Not-a-Neg]

It launched with embedded OEM systems in December.

Power savings?

[Anonymous Coward]

Are they still going to draw 220 watts like the FX-8350?

Re:Power savings?

[Anonymous Coward]

93 Watts from what they have stated for that.

Re:Power savings?

[jwhyche]

The FX-8350 is rated at 125W. You are thinking of teh FX-9950.

Re:Power savings?

[Anonymous Coward]

Slightly less than Intel at slightly better performance, if you believe the demo.

Too bad I shot you down easily... apk

[Anonymous Coward]

See subject: Where YOU failed trying to tell ME "how to write code" AGREEING w/ me on my design https://slashdot.org/comments.pl?sid=10053471&threshold=-1&commentsort=0&mode=thread&pid=53577421/ + https://tech.slashdot.org/comments.pl?sid=10053471&threshold=-1&commentsort=0&mode=thread&pid=53567665/ which others like & use here (many 1,000's worldwide w/ them) THEN telling me "depend on OS memmgt & processscheduling" which MS has FUCKED UP ROYALLY BEFORE on (timeslicing in code) https://slashdot.org/comments.pl?sid=10053471&cid=53585435/ & (memmgt) https://slashdot.org/comments.pl?sid=10053471&cid=53588061/

* You had best work on your rep (& I've been using MY INITIALS far longer than you OR EVEN GOOGLE) after those bungles of yours...

(Lies too? I don't START flamewars - I finish those starting them w/ me, easily, just as I did you as seen above).

APK

P.S.=> How could I forget I showed you the SINGLE place my code (mostly linear blocking process - can't proceed, like a Gannt chart works, unless priors finish) DOES do multhreaded concurrent ops perfectly https://tech.slashdot.org/comments.pl?sid=10053471&cid=53570565/ - you're ALL talk & evidently you cannot THINK for yourself (so don't even TRY think for me, you don't have the mental chops or coding skills)... apk

Re:Too bad I shot you down easily... apk

[NotAPK]

OK mate, I've read enough of your BS to know there's no point arguing with stupid people: they beat you down to their level and then win from experience.

Have a nice life...

Re:Too bad I shot you down easily... apk

[Anonymous Coward]

So, only one part of your code can do multithreaded concurrent operations?

Son, Quake 3 did that decades ago and that was roughly HALF of the codebase. Why the fuck are you behind the ti-oh wait, you advertise HOSTs as a security solution, nevermind, proceed to be out of date while real people run hardware solutions.

misspelled

[TsuruchiBrian]

Wasn't it ment to have been launched in 2016?

Wasn't it supposed to be spelled meant?

Do you see what I did there? Did this add anything to the discourse?

Re:misspelled

[Anonymous Coward]

Wasn't is supposed to be 'I like a good dick in my ass?'

Do you see what I did there? Did this add anything to the discourse?

Let's not hallucinate

[Master5000]

Intel is a monopoly. They have 10 times more money than AMD. Do you really think AMD has a chance?

Re:Let's not hallucinate

[Anonymous Coward]

However, this time there's a precedent. If Intel tries the same shit they did the last time AMD had them on the ropes, I doubt they will get away with it anyway near as cheaply.

Re:Let's not hallucinate

[Bryan Ischo]

Why, yes I do. Simply put, all the low hanging fruit x86 processor development has been picked already. The only remaining advancements are incredibly expensive to make, both in time and money, and as a result even Intel can't appreciably speed up x86 processors anymore.

Now that Intel is mostly standing still, AMD has a chance to catch up. It can benefit from all of the research work, process developments, etc, that have occurred, and implement its own version of the same advancements that Intel made over the past 7 years or so, much more cheaply than Intel did. AMD's limited budget may just be enough.

I actually don'tt think that Ryzen will be as fast as Intel's best. But I expect it to come within 25%, and for a reduced cost. My prediction is that x86 won't get much faster as AMD becomes more competitive with Intel, but it will become a lot cheaper. Intel's high end consumer chips that cost $350 now will be available for $100 in a couple of years, whereas without AMD's Ryzen providing the competition, they probably would only have come down by $50 or so.

Re:Let's not hallucinate

[AHuxley]

It just has to be as good as an i7 and not cost as much.
Games like a few cores, as long as each can keep up with the best AMD and NVIDIA gpu's.
The ability to add a huge cpu cooler and let the CPU ramp up will be interesting.
As long as one or more GPU's get the CPU support they need for the most demanding 4K ready games vs an i7/i5 it will be fine.

Re:Let's not hallucinate

[Anonymous Coward]

Intel is a monopoly. They have 10 times more money than AMD. Do you really think AMD has a chance?

Considering how many fabs Intel has to support I should hope they have 10x the cash of AMD. There is still plenty of money to go around in x86 and you'll know when that changes by Intel consolidating its fab operations.

You're all BS 'talk' vs. facts I use... apk

[Anonymous Coward]

See subject: I've written a good program (many here like & use it + malwarebytes' folks host & recommend it) you? Not that I can see... ALL I see from YOU, is CHEAP TALK (nothing more) I blew away, with ease (& documted facts from reputable sources).

APK

P.S.=> I also see that you can't validly technically disprove the process is MOSTLY linear & that I do indeed use threads where appropriate (backgrounded favorite sites hardcoded)... apk

Re:You're all BS 'talk' vs. facts I use... apk

[Anonymous Coward]

What I have learned? Anyone with the username APK is probably a fucking dumb shit who deserves to be ignored.

Re:You're all BS 'talk' vs. facts I use... apk

[Anonymous Coward]

I disagree. NotAPK is a fucking dumb shit who deserves to be ignored after APK annihilated him https://hardware.slashdot.org/comments.pl?sid=10106903&cid=53650085/

More than your non-existent code, lol

[Anonymous Coward]

See my subject above? Hahahahahaha @ YOU, talker!

* Hot windbag BLOWHARD bs = zero to me (& anyone else)

APK

P.S.=> I also see that you can't validly technically disprove the process is MOSTLY linear & that I do indeed use threads where appropriate (backgrounded favorite sites hardcoded)... apk

As an AMD fan since 1995

[Not-a-Neg]

I am more excited by the use of a single socket (AM4) for the entire platform. It will now be possible to use their non-APU processors on an ITX board which previously have been limited to the FM# sockets. Don't need Intel overpriced crap for an extra 5FPS.

"Hardware 'solutions'" BLOW (bugridden)

[Anonymous Coward]

See subject above & proof thereof (far from complete yet 100's of router problems) https://it.slashdot.org/comments.pl?sid=9995967&cid=53488785/

* So where's your code that does as much as mine does from what you already have that operates in kernelmode (vs. slower usermode & inefficient as hell + bug ridden (DNS/antivirus) or crippled NOT to work by default (adblock)) giving you more security, speed, reliability, & anonymity for less (less bugs + resources consumption)? Forget multithreaded (which I do easily) - just code that does all I noted just now?

APK

P.S.=> ANSWER = it's not - you're just a HOTAIR windbag unskilled LOW troll, nothing more (& you KNOW it) - funniest of all? I'll concede to a FELLOW /.'er in John Carmack (he's great & one of my 'tech/intellectual heros' in fact) - YES, he's better than I by far - but the "troll likes of you" BLOWHARD? No way, lol - you're ALL 'talk' & talk's cheap chump... apk

"You're a smart guy APK"

[Anonymous Coward]

See subject & those very words quoted from YOU:

"You're a smart guy APK" - by NotAPK ( 4529127 ) on Saturday December 31, 2016 @12:41PM (#53585041)

... & yes you said that: I'm only ~ 130-135 IQ, & I don't consider myself 'smart' - John Carmack's smart (one of my tech heroes) - I can "get the job done" though... & I do!

APK

P.S.=> Plus, hey (lol, I gotta do this): IF that was the case (stupid people beating you down to their level & beating you with experience)?

THEN YOU OUGHT TO BE BEATING ME!

(Vs. me doing a "cardinal richelieu" on YOU letting you "EAT YOUR WORDS", beating yourself - & the rest was my dismantling you technically here easily https://hardware.slashdot.org/comments.pl?sid=10106903&cid=53650085/ which hey, again: YOU DID THAT TO YOURSELF...apk