AMD Announces Ryzen 5 Processors With 4 and 6-Core Chips Starting At $169 (hothardware.com)

MojoKid writes from a report via HotHardware: Today, AMD unveiled additional details with respect to the entire Ryzen 5 processor line-up. Unlike the Ryzen 7 series, which consists entirely of 8-core/16-thread processors, the Ryzen 5 family has two tiers consisting of 6-core/12-thread and 4-core/8-thread processors. The entry-level part is the Ryzen 5 1400, a 4-core/8-thread CPU with base and turbo clocks of 3.2GHz and 3.4GHz, respectively. The Ryzen 5 1500X has the same quad-core configuration, but with base and turbo clocks of 3.5GHz and 3.7GHz, and also has support for an extended XFR frequency range of up to 3.9GHz. The Ryzen 5 1600 is a 6-core/12-thread processor, with 3.2GHz base and 3.6GHz boost clocks. And at the top of the stack is the Ryzen 5 1600X -- which has a similar 6-core configuration -- but cranks things up even further to 3.6GHz/4.0GHz. With XFR, the absolute maximum frequency for all of the Ryzen 5 processors will be somewhat higher, but AMD hasn't disclosed specifics for all parts. AMD's Ryzen 5 processor line-up will work with the very same AM4 platform as the higher-end Ryzen 7. Ryzen 5 series processors will be launching officially on April 11, with prices starting at $169 for the Ryzen 5 1400. An additional $20 will get you a Ryzen 5 1500X, while the 6-core Ryzen 5 1600 and 1600X will sell for $219 and $249, respectively.

Comments

Intel

[Anonymous Coward]

Is in real trouble if they don't turn around their operation, quickly

Re:Intel

[Anonymous Coward]

LOL. 20 years ago people were saying the same thing about the K6.
Since then AMD and Intel have swapped performance leads dozens of times.

tl;dr: I don't think either company is worried -- especially since they're not competing for the same market.

Re:Intel

[Anonymous Coward]

Intel have been selling their CPUs at highly inflated prices for years, just because they could, and people don't give a shit that they've been literally robbed. Hardly at risk, but maybe one day people wake up and realize it.

Re:Intel

[supremebob]

By "trouble", you mean having to decrease their CPU prices slightly more than expected over the next few quarters, sure. AMD can only dream of having profit margins that equal Intel's right now.

Re:Intel

[unixisc]

Between then and now, AMD acquired a part of the DEC Alpha team, and had Dirk Meyer as their CEO. They did well w/ the Athlon, and later achieved a coup w/ the AMD 64 architecture, by getting Microsoft's endorsement. This was at a time when Intel was struggling w/ the Pentium 4, and the Itanium was such a disaster that Intel had to follow AMD's lead.

But AMD never grabbed the initiative to build on their fab capabilities and manufacturing processes, instead continuing to focus just on low cost CPUs. Never a winning strategy, since not only did Intel have pockets deep enough to price match them anywhere they chose, they also had a 2-3 generation advantage in terms of process nodes. Ultimately, AMD threw in the towel and sold off their fabs to Global Foundries.

And unlike the time when AMD acquired another great CPU team, this time, there are no CPU teams left to acquire. All that there is in the market are Intel and ARM. For ARM, there are all the major licensees - Qualcomm, Phillips, TI, Samsung, et al, so there are no great teams they can really acquire. Despite the occasional flashes in the pan, AMD is pretty much where Cyrix, Winchip and Centaur are - swimming w/ the fishes

Re:Intel

[Anonymous Coward]

Literally? They were literally robbed? Are you sure you didn't mean the exact opposite?

Re:Intel

[Rockoon]

But AMD never grabbed the initiative to build on their fab capabilities and manufacturing processes

Thanks to Intels monopolistic practices. They were convicted of it on multiple continents.

instead continuing to focus just on low cost CPUs.

Wrong. Possibly intentionally.

Never a winning strategy, since not only did Intel have pockets deep enough to price match them anywhere they chose, they also had a 2-3 generation advantage in terms of process nodes.

Now we know you are lying.

Ultimately, AMD threw in the towel and sold off their fabs to Global Foundries.

AMD didnt sell their fabs. They spun off a new company. More proof that you are a liar, and the only motive here for lying is because you are a monopoly apologist that didnt mind getting fucked by Intel bribing companies to only sell their worse technology for higher prices.

Re:Intel

[ArchieBunker]

About the only thing left is the POWER series that IBM is pushing. I'd really like to see a desktop computer based on the AS/400 architecture. So different and much more logical. Tons of low level stuff normally handled by the OS is done in hardware.

Re: Intel

[Anonymous Coward]

The expression is "sleeping with the fishes," a reference to being killed (typically murdered) by drowning.

Swimming with the fishes implies that everything is going fine. Which makes no sense.

Re: Intel

[Anonymous Coward]

Literally meant to say "virtually robbed"

Re:Intel

[Anonymous Coward]

You can try to blame Intel for all of AMD's problems but AMD's absolute shit performance since bulldozer is entirely AMD's fault.

The man in charge fired all those great devs they got from the Alpha team and decided R&D was a cost center. - In what is probably the most high-tech industry in the world this was a dumb move and is 110% AMD management bungling. And just the start.

AMD has really been badly mis-managed for the last decade and only recently has gotten back on track.

AMD was complete shit and wit was AMD's fault. Stop fanboying and whining about things Intel did decades ago. It's 2017.

Re:Intel

[TsuruchiBrian]

I'm not sure AMD will want Intel's profits now that they will be forced into a price war. Intel has a lot of money, but I don't see much profit in the immediate future for them, at least not the kind of profit they enjoyed for the last ~decade with 0 competitors.

Re: Intel

[Anonymous Coward]

The expression is "sleeping with the fishes," a reference to being killed (typically murdered) by drowning.

Swimming with the fishes implies that everything is going fine. Which makes no sense.

Humans are capable of voluntary control of their breathing allowing temporary suspension of the inhale/exhale cycle while continuing cellular respiration on air stored in their lungs, but cannot maintain that control while unconscious so a sleeping human will drown if submerged but a conscious human can safely swim for a time dependent on physical condition and endurance.

Re:Intel

[unixisc]

You are totally clueless, and it shows:

1. How would anything Intel did prevent AMD from investing in their own fabs? The ones they had were anaemic - and while they attempted some good ones, like Dresden, their process methodologies never advanced much beyond their basics

2. They did decide to focus on CPUs that would beat Intel in price, ignoring the huge advantage Intel had in cost. Instead, had they focused more aggressively on next gen process nodes and improving their yields, they'd have done a lot better

3. Intel is widely recognized as a leader in fabs, ahead of the best of them - previously IBM, TSMC, Samsung, et al

4. Whether they sold or spun off is a distinction w/o a difference: bottom line is that AMD has less control on their supply, and Global Foundries is at liberty to give them a lower priority if they get better customers

Intel's technology is worse? Only in some alternate universe

Re:Intel

[unixisc]

But IBM already exited the semiconductor market a while ago - in fact, they too outsourced it to Global Foundries. Hardly much of a volume driver, which is what any semiconductor house needs in order to be able to recoup its investments quickly.

Re:Intel

[unixisc]

Intel is pretty much the best that's left of manufacturing in the US - most R&D, most innovative and headed by someone from their fabs. In the worst case situation - assume that the market for x86/x64 completely dries up, they can simply be a US version of a TSMC/GSMC, and still be above water.

Most companies would kill to be in that predicament. AMD, OTOH, has nothing left but the x64, and its hybrid ARM/x64 lines look interesting, but it's not much more valuable than any of the myriad number of fabless semiconductor companies out there. They are behind Intel in a 2 horse (or a horse vs donkey) race in CPUs, and they are behind NVIDIA in a 3 horse race in GPUs. Every other business they once had - be it FPGAs, Flash, Embedded CPUs, et al has been closed or sold, so there's not much else they have, really

Re:Intel

[TsuruchiBrian]

assume that the market for x86/x64 completely dries up, they can simply be a US version of a TSMC/GSMC, and still be above water.

Which would mean that the part of Intel that is analogous to what AMD is now would be dead along with AMD, and the part of Intel that is analogous to GSMC would still be around along with GSMC

Would it really make any difference in that scenario if GSMC was still called AMD?

Re:Intel

[samoanbiscuit]

But AMD never grabbed the initiative to build on their fab capabilities and manufacturing processes, instead continuing to focus just on low cost CPUs.

Fabs cost money. It's not about initiative, but the ability to pony up billions for several years. Billions they never got because Intel bribed the major OEMS at the time to refuse to put AMD in their products, or to include them only in their shittiest products, and at small volumes. Even when the Ahtlon 64 was eating Pentiums for breakfast. Dell's existence was basically subsidized by Intel during these years thanks to the bribes

During the several years AMD had the performance crown, they were largely relegated to dumpster tier OEM products or custom built enthusiast PCs (not the largest market segment) due to Intel's illegal business practices. The 1 billion fine Intel was eventually slapped with is pocket change compared to the opportunity cost of the revenues they could have earned during their time in the sunshine, and how they could have used those missing dollars in R&D, operations and marketing.

they also had a 2-3 generation advantage in terms of process nodes. Ultimately, AMD threw in the towel and sold off their fabs to Global Foundries.

Intel only pulled ahead of AMD in the process node race at ~32nm, years after they fucked AMD over financially. And process node shrinks are a function of how much money you can throw at the problem. Since the smartphone explosion lead to major money and mindshare being thrown at third party fabs, TSMC and Samsung are quickly catching up to Intel, and look to be on the verge of surpassing Intel in the next few years. Intel's lead was once thought insurmountable, now these upstarts are snapping at it's heels.

And unlike the time when AMD acquired another great CPU team, this time, there are no CPU teams left to acquire.

AMD's hardware teams are easily among the best in the industry, high profile hires are just one part of the story. Intel and nVidia engineers readily acknowledge that fact all the time. Under much more resource constraints than the two incumbents, they regularly put out hardware that challenges and even beats what the two industry behemoths put out. Their driver/software people are a different story, but that's understandable when you compare AMD's workforce (~9000) to Intel (`100,000) or nVidia (~10,000).

Re:Intel

[sixsixtysix]

They did bring Jim Keller back for this one.

This seriously gives me a boner

[Anonymous Coward]

Excuse me while I whip up some "thermal paste".

TDP?

[phorm]

It's good to see a broad selection of cores/threads and clock coming out, but what I'd be really interested in knowing what the actual wattage/TDP of this processors is supposed to be.

I've had a mini-ITX box running with an A10 and a slotted GPU which can pretty much hold its own for any games etc, but I would like to get something a bit more powerful or more cores. Normally you're not going to be able to run a really high-wattage CPU on a mini-ITX board, and even if you could the tight spaces tend towards overheating. It would be nice if the 6-core CPU's can balance out with a reasonable wattage, and even nicer would be if they come out with some E-Series CPU's (great Performance Per Watt on those) under the Ryzen platform

Re:TDP?

[K. S. Kyosuke]

Price/performance (and power/performance) wise, you're probably still better off with an 1700.

Re:TDP?

[Anonymous Coward]

They did mention it in the article. 1600X is going to be 95W, the rest - 65W

Re:TDP?

[Anonymous Coward]

I was expecting less, given that the Ryzen-7 comes also in 65W (1700) and 95W (all the others) versions.
I have to build soon a new computer to replace my over 10 years old Athon64-X2, and I was expecting the quad core to be, not a power miser, but well below 50W TDP.

Re:TDP?

[OpenSourced]

I'd be really interested in knowing what the actual wattage/TDP of this processors is supposed to be.

Yeah, if only there was a way of knowing that, like for example reading the linked article!

Re:TDP?

[Anonymous Coward]

nobody has time for that luxury!

Re:TDP?

[phorm]

OK. I see that it is actually listing them as 65W and 95W CPU's. My bad for searching for the term "watt" and "tdp" instead of a specific number.

Re:TDP?

[aliquis]

Worth noticing is that the 95 watt Ryzen 7 1700X and 1800X increase the power-demands more when going from idle to fully loaded than the 140 watt i7 6900K as well as the whole platform using slightly more power for the whole system:
http://www.sweclockers.com/tes...
So take the TDP values with some salt. It doesn't tell the whole story. Neither have they done for AMD vs Nvidia graphics cards I've heard claimed but there with the advantage for AMD.

Ryzen 7 1700X and 1800X idle whole system at the wall: 45 watt
i7 6900K idle whole system: 60 watt.

1700X blender & x264 benchmark: 155 & 160 watt.
1800X blender & x264 benchmark: 165 & 170 watt
i7 6900K blender & x264 benchmark: 150 & 150 watt.
That's still for the full system.
i7 6900K Intel Ark says 140 watt TDP: https://ark.intel.com/sv/produ...

i7 6950X: 60/145/170 watt idle, blender, x264 (10 core also 140 watt.)

Re: TDP?

[Anonymous Coward]

Well AMDs thermal envelope is half of Intel's with Ryzen. Not to mention 3x treads for the same price.

So, it would be really retarded move to buy Intel atm.

Re:TDP?

[Anonymous Coward]

Many of the new Ryzen chips have unlocked multiplier, so can probably get theme very close to Intel's power consumption by overclocking.

Re:TDP?

[Anonymous Coward]

Low idle power usages are nice for mobile parts but all you're really saying is AMD cpus are better when they're turned off.

Re:TDP?

[PRMan]

But do you run your computer crunching stuff 24/7 or does it sit idle most of the time?

Re:TDP?

[Anonymous Coward]

That's not the point. The point is that the advertised numbers are fiction, which makes it harder on people who are trying to plan carefully around power supplies and cooling.

If I get a Ryzen part - and I'm generally an AMD fan, so I probably will - I've seen enough to get a good cooler and plenty of extra capacity in the power supply. I can't assume '65 watt TDP' listed next to the R7 1700 has any useful meaning.

Re:TDP?

[Anonymous Coward]

Not cheap, but I run an 8 core/16 thead 45 watt intel D1541 chip.and 64G of memory. Its great for virtualization.

Re:TDP?

[aliquis]

No, I'm saying power usage increase over 120 watt with the 95 watt TDP CPU and less than 100 watt with the 140 watt TDP CPU.

And that the 95 watt TDP CPU system use more power during load than the 140 watt TDP CPU system.

What you claim is something you made up yourself.

Re:TDP?

[bongey]

It's Game over for Intel in the server market on power consumption. A slightly under-clocked Ryzen 1700 scores 850 in cb, and draws only 30 Watts at full load. Intel's low power offering Atom c2000 CPU's draw 33-35W under full load. Intel really will loose 15-30% new server chips sales on this alone, Xeons are 90-140w under full load.

Re:TDP?

[bongey]

Forgot link https://forums.anandtech.com/t... . I have Atom c2750 .

Re:TDP?

[Shirley Marquez]

Depends on your usage. The 1600X looks like a better gaming CPU than the 1700 and it's effectively $50 cheaper. (The list price is $80 less but no cooler is included.) On the other hand it will consume more power at peak; the 1600X is a 95W TDP processor while the 1700 is 65W TDP. We won't know what the idle power story is until the Ryzen 5 CPUs are released and benchmarked.

Looks like the i5-7600K and i7-7700K have some serious competition ahead. The AMD parts will cost less than Intel's, even after the recent cuts in street price of those CPUs, and AMD motherboards are also less expensive. By the time they're available, the scheduling issues that are holding back Ryzen's performance in games should be resolved.

Re:TDP?

[Shirley Marquez]

If keeping down to the 45W TDP level or even lower is a priority you can underclock. But CPUs optimized for really low power consumption will come later, and I suspect they will be APUs rather than processor-only parts. A 15W Zen-based APU will be awesome for applications like HTPCs.

Re:TDP?

[lsatenstein]

For normal youtube, browsing and games, published value TDP 65watts for the 1600x If you go for high frame rates, and super busy graphics, the likely hood is that the wattage could hit 80Watts

processor line-up

[Anonymous Coward]

Is this "processor line-up" really just one processor with random defects, which gets tested for correctness and speed, and then binned based on the results?

Re:processor line-up

[K. S. Kyosuke]

That's usually how it works these days. Intel even switches off entire perfectly working features. AMD at least does it much less so.

Re:processor line-up

[Anonymous Coward]

So the absence of a 5-core configuration in the announcement means that AMD sometimes switches off a perfectly working core?

Re:processor line-up

[K. S. Kyosuke]

Presumably, many 6-core chips are going to be disabled 8-core chips, therefore with two cores disabled. Pretty much for the same quantitative effect as i5s being partially disabled i7s.

Re:processor line-up

[Highdude702]

Thats because Intel doesn't actually give a fuck about its consumers unless you have a net worth of at least $1 million dollars...

Re: processor line-up

[Anonymous Coward]

Ever. That's how it's worked ever since there have been speed-priced CPUs. The whole industry has done this forever.

What you do occasionally see is too-high yields on, say, 2GHz parts and the manufacturer is forced to label them as 1.8GHz to fill existing orders. That doesn't happen as often, but it does happen.

Re: processor line-up

[Jon Peterson]

Indeed. The Intel 486dx 66Mhz (early 1990s) was simply a 486dx 100Mhz that had failed a QA test at that speed and was re-tested at 66Mhz and then sold if it passed.

It's an efficient way to deal with inherent fragilities of manufacturing at the limits of technology.

That said, market segmentation whereby you make one product and sell disabled versions of it at different price points has been going on even longer. The economies of scale make it cheaper to do this than to make physically different products for each segment. VCRs in the 80s and 90s were made like this, such that they all had the same internals, and the difference in models was achieved in external styling and what buttons were made available, and what firmware was installed.

Re: processor line-up

[Anonymous Coward]

I have a net worth greater than that and they don't care about me, at all.

Re: processor line-up

[drinkypoo]

VCRs in the 80s and 90s were made like this, such that they all had the same internals, and the difference in models was achieved in external styling and what buttons were made available, and what firmware was installed.

Sure, some of them were. But they also still made play-only heads into the 90s. You couldn't necessarily turn a VCP into a VCR. I'd guess until 1995 when the heads changed anyway.

Re:processor line-up

[gman003]

The Zen architecture uses four-core "Complexes", which combine cache and some cache-coherency logic (not execution resources, the way Bulldozer did). The six-core parts are confirmed to be two-complex (eight-core) parts with two cores disabled. The four-core parts are almost assuredly single-complex parts, ie. a die specific for four-core (and under?) chips.

Re: processor line-up

[Highdude702]

Those dirty cunts!

Re: processor line-up

[Highdude702]

Im going to tell you what I tell my father, who from what you said im assuming is rather close to you in age... SHUT UP OLD MAN! :)

Re: processor line-up

[Anonymous Coward]

are you mad bro?

Re: processor line-up

[Anonymous Coward]

The DX4 you refer to came out a year later than the DX2 that you imply is the downgraded types.

Re: processor line-up

[Anonymous Coward]

The Intel 486dx 66Mhz (early 1990s) was simply a 486dx 100Mhz that had failed a QA test at that speed and was re-tested at 66Mhz and then sold if it passed.

While I agree with your overall point, the example you used here doesn't support it. The first Intel 486 running at 66 MHz was released in 1992 (the 486DX2). Intel didn't release a faster 486 until 1994 (the 100 MHz 486DX4). The 66Mhz 486DX2 was Intel's flagship 486 for 2 years after its release, it wasn't a marked-down unit that failed testing.

Re: processor line-up

[Anonymous Coward]

Nope. The Intel 486DX2 was a successor to the 486DX. While AMD had pushed its 486 design to 40 Mhz, this was causing clock issues on the external buses. And Intel's 50 Mhz design was causing heat problems. The Intel solution (which we use to this day) is to have a clock multiplier - the internal clock would run at a multiple of the external clock. And for their initial design, that multiplier was a fixed factor 2. Hence the DX2, running internally at 2x33 = 66Mhz - significantly faster than the AMD at 40 Mhz.

The DX4 was the last design before the Pentium, and the first to get funky with names. This was the design you're thinking of, running at 3x33 Mhz. And it's speed-binned variant was the DX4-75, not the DX2.

Re: processor line-up

[PRMan]

I got a "2-core" AMD "Sempron" which was really just a Phenom with 2 cores turned off. I turned them back on on my motherboard and ran 4-cores the whole time.

Re: processor line-up

[PRMan]

Yep. At first. And then they all worked so you could easily buy a cheap 486/66 and run it at 100 no problem.

Re: processor line-up

[radarskiy]

No one has fused off working features in over a decade. Every CPU manufacturer is supply constrained at the high end: if they had more top bin parts, they could sell them at top bin prices.

These days if a feature is fused off it is because it failed and the manufacturer is attempting die recovery.

Re: processor line-up

[Shirley Marquez]

GPUs also do this. A lot of the model differentiation in video cards involves selling of cards that are partially disabled and/or downclocked because they failed to meet specifications with everything turned on. They may have some completely broken parts, they may have failed at the full clock speed, or they may have consumed too much power at that speed. In some cases they actually passed all the qualifications but are sold as lesser (and less expensive) parts because of lack of demand for the most expensive model. When you buy there is no way to know which of those things you got, though you can run tests at home to try to figure it out.

It also happens with parts that don't contain processing power. DRAM comes in a variety of speed grades but they all come from the same fab line; they get sold with various speed ratings (and prices) based on how well they perform when tested.

Re: processor line-up

[lsatenstein]

So in all likelihood, your 1600x could be a crippled or "injured" 1800x that was put into the 1600x package

Re:processor line-up

[Shirley Marquez]

Initially, these are probably all going to be eight core die with some cores disabled. But there are some interesting performance implications depending on what parts of the chip are disabled. It's entirely possible that different six and four core chips of the same model will not perform identically.

At the six core level there are two possible configurations: you could have one where one of the four cores of each complex is disabled, and another where one complex is fully enabled and the other has two cores turned off. Each type would require different process scheduling to perform optimally.

At the four core level you could have chips where one complex is turned off entirely, and also parts where both complexes are partly active. The low end Ryzen 5 1400 has only half as much L3 cache, so it's almost certainly the first type and runs only one complex. The 1500X has the full 16MB of L3 so it's presumably the second type. That could be either 2/2 or 3/1 and each would have to be scheduled differently.

Frankly this is what they should have launched 1st

[Sycraft-fu]

The current Ryzens are kinda underwhelming. Not because they are bad chips, but I find myself in a position where they wouldn't be what I'd recommend to most people.

For your average user, they are way overkill. Hell even a quad core is overkill for normal desktop/media consumption/etc usage. They are too expensive, a cheaper Intel i5 or i3 is the way to go.

For gamers, they don't perform as well as Intel's high end in general. Games are multi-threaded these days, but generally have one main thread that is the big limit and the others are much smaller. It is rare to see them max a quad core, much less more than that. For best performance they need high clocks and IPC, and the 7700k does a better job at that for about the same money as the 1700. Only 4 cores, but as I said just doesn't matter to games. Likewise the 7600 or 7600k are in general better and even cheaper.

For audio production Ryzen seems to have higher latency. Maybe this gets resolved later, but right now you need to set your ASIO buffers higher to avoid dropouts with the same project. Also since CPU load isn't generally the limiting factor (CPUs are very fast compared to the needs of audio processing) the extra cores aren't useful unless you do REALLY heavy mixes. So better to go for an Intel CPU and get lower latency for cheaper, or maybe an Intel HEDT CPU for the same price as the 1800X.

Video encoding is the one area they seem to really win at. There the more core equal more performance and you can get 8 for the price Intel sells you 6. So if that's what you are after, then it is a good deal. Not really the most common use.

These though, for the price they should be killer. The 4 cores are likely to at least compete with the 7600 and cost you a good bit less. Could make them a very good contender for gamers, or just general desktop users that want a solid system.

I'll be real interested to see the benchmarks when they come out.

Re:Frankly this is what they should have launched

[K. S. Kyosuke]

Programmers are going to be very happy with affordable 8-core chips, though. I know I will.

Re: Frankly this is what they should have launched

[Anonymous Coward]

How came? Maybe if I stuff that thing with 32GB of RAM and a couple SSDs in RAID...

Re:Frankly this is what they should have launched

[calken1979]

It will be 6 weeks after the initial Ryzen 7 launch - you make it sound like it was 6 months. Besides, with the Ryzen 5 CPUs most likely being faulty 7's, it takes time to build that stock level depending on yield. I can see the Ryzen 3 being just a single core complex with the perfect samples moving into the Ryzen 5, 4 core range and the basic 3's having the SMT turned off.

Re: too little too late

[Anonymous Coward]

Man, those are different kinds of cores. Multi core butthole vibrators are still a novelty in most countries, people seem to be just happy with regular vibrators, because application areas don't benefit much from having more than one core.

And remember, bigger more powerful cores are more suitable for single task applications.

Crippled Ryzen 7

[Misagon]

Unfortunately, it seem as if these 6-core and 4-core Ryzen 5 CPUs are only going to be eight-core Ryzen 7 CPUs with cores disabled in both compute-complexes.

The R5 1600X and 1600 are going to have one core disabled per compute-complex (CCX): 3+3. This was expected.
However, surprisingly, AMD has told Anandtech and Ars Technica that the R5 1500X and likely also the R5 1400 are going to have two cores disabled per CCX: giving it a 2+2 config.

When clock and IPC have been taken into account, Ryzen's biggest performance bottleneck compared to Intel has been shown to be when threads on different CCX'es are accessing the same memory. Each CCX has its own L3 cache and there is an interconnect between the CCX'es L3 caches which while being slower than a single shared L3 cache is somewhat faster than going to main memory ... but the L3 caches are only victim caches to each core's L2 cache - and therefore not necessarily caching the entire working set.

This means that the 1500X and 1400 are going to be slower on many workloads than on a hypothetical Zen CPU with one single four-core CCX.

It is believed that this bottleneck is the reason behind relatively low Ryzen 1800X/1700X/1700 scores in many games - compared to Intel (even when clock speed and IPC have been taken into account).
(Curious enough, this is also a known issue among programmers for the XBox One and PS4 - both having AMD CPUs with a similar setup, but apparently it didn't really occur to game programmers that AMD would have a go at retaking the desktop?)

Re:Crippled Ryzen 7

[Anonymous Coward]

well, it is much cheaper and more efficient for them to just fab one part, and bin based on quality checks or demand. sucks when TDP usually remains higher and as you said, some cache access is slower than a redesigned 'smaller' chip, but what can you do? not much. nothing. spend more on intel? who's gone so far as to have previously sold unlocks for disabled cpu features like a game DLC? fuck no.

Re:Crippled Ryzen 7

[aliquis]

Unfortunately, it seem as if these 6-core and 4-core Ryzen 5 CPUs are only going to be eight-core Ryzen 7 CPUs with cores disabled in both compute-complexes.

According to what? TDP numbers?
I would hope the 4 core was just one CCX.
For 6 core I don't know if I think 3+3 or 4+2 is better, for gaming I'd take 4+2 and then the +2 could be used for other threads.
The split cache isn't 100% useless. Assuming the same bandwidth to it (not a fact or even thing I think) having just half the cores use it could improve performance to it, the very weak connectivity between them however is an issue. In reality the latency within the same CCX is lower than over at Intel but between all cores on average is worse and to the second CCX from the first and viceversa is even worse.

Seem like I should had read your whole comment .. ;D

The R5 1600X and 1600 are going to have one core disabled per compute-complex (CCX): 3+3. This was expected.

Too bad. Still 16 MB of L3 cache or disabled so one get the worst of everything? :D

However, surprisingly, AMD has told Anandtech and Ars Technica that the R5 1500X and likely also the R5 1400 are going to have two cores disabled per CCX: giving it a 2+2 config.

Ok.
I wonder if that's why they went with Overwatch which only use 2 cores:
https://www.youtube.com/watch?...
It's still a good processor for the price though. For many tasks due to having the 6 cores it will be more powerful than the competing i5 7600K. VS an overclocked one it will likely be a bit slower in some games of today, non-overclocked maybe not / not much, but it still have those extra cores for background tasks and streaming and possibly future advantage. Windows 10 game mode will also dedicate some cores for gaming and prevent threads from moving around so that will bring some more performance.

and there is an interconnect between the CCX'es L3 caches which while being slower than a single shared L3 cache is somewhat faster than going to main memory

I read before that it was on the same bus as the PCI-express lanes and the DDR4 RAM. Correct or not?
I've also read it ran half speed of the RAM speed (in clock?), but it seem weird that the bus / cache would vary speed with what RAM one use. However I've seen different L3 latency test numbers in different memory configurations, thought that could be the benchmark software which happen to access some regular RAM too I suppose.

This means that the 1500X and 1400 are going to be slower on many workloads than on a hypothetical Zen CPU with one single four-core CCX.

Yeah. I would prefer that and it would feel better if they just made 1 CCX chips for that purpose .. ... then again maybe they let people unlock functional cores?! ;D (but fuck up overclocking when doing so? ;D .. reason to wait? ;D)

It is believed that this bottleneck is the reason behind relatively low Ryzen 1800X/1700X/1700 scores in many games - compared to Intel (even when clock speed and IPC have been taken into account).

Both cache and memory performance is slight worse. It still have just 2 AGUs / core just like FX before it whereas Intel have 4, I don't know how much that matter or affect things (It also only capable of 2 128 bit FMAC vs Intel 2 256 bit FMAC and hence is inferior there too, I don't know how much that matter. To me it would make sense if software which relied heavily on floating-point performance used the GPU instead (though high precision performance on consumer GPUs are most often much slower than single-precision so maybe they consumer graphics cards isn't better there than an Intel HEDT/Xeon processor anyway?) but I also assum

Re:Crippled Ryzen 7

[Gravis Zero]

Unfortunately, it seem as if these 6-core and 4-core Ryzen 5 CPUs are only going to be eight-core Ryzen 7 CPUs with cores disabled in both compute-complexes.

If that's true then they are probably tweaking their design doing that to increase their production yields. QA is important when it comes to making complex chips and when you can identify issues before they become returns then you can turn them into sales instead of a loss. Sony did the same thing for the PS3 processor by disabling one core. This is actually a good practice to perfect your chip. However, beware of chips with "locked cores" because that's just pure exploitation.

Re:Crippled Ryzen 7

[supremebob]

On the flip side, some smart engineer is going to figure out how to re-enable those disabled cores through a firmware hack or some other trick. At that point, you'll be able to get the equivalent of a $400 CPU for $150.

Re:Crippled Ryzen 7

[hairyfeet]

Actually this could work to AMD's favor if they allow one to take their chances and unlock disabled cores like they did with the Phenom IIs and Athlon IIs. I know a LOT of guys that bought those chips simply so they could gamble on getting a better chip for the money, after all at the worst they would get the chip they paid for and if they got lucky they would get 1 or 2 cores for free, there was even a Phenom II quad that you could not get for the longest time after it was leaked it was a hexacore with 2 cores disabled as everyone was buying them to take a shot at a free hexa upgrade!

So if they do a soft disable like with Phenom II? Man I can see them selling a ton of those chips as everyone rolls the dice, after all who doesn't like the thought of taking a shot and getting a free upgrade?

Re:Crippled Ryzen 7

[Khyber]

Pretty much every fab laser-locks that out now days.

Re:Crippled Ryzen 7

[thegarbz]

Unfortunately, it seem as if these 6-core and 4-core Ryzen 5 CPUs are only going to be eight-core Ryzen 7 CPUs with cores disabled in both compute-complexes.

There's nothing unfortunate about this. By crippling the low end hardware they increase the number of units sold thanks to their lower price. This in turn pays back R&D costs that would otherwise be spread over a smaller number of sales making the high end even more expensive.

Re:Crippled Ryzen 7

[epine]

I'm still waiting to discover the on-chip interconnect. I'd imagine that some kind of on-chip TSV would provide a hefty upgrade in the width of the data path between the CCX modules, that would still have some penalty, though far less than the historic standard.

I really can't see how Naples is going to use the same CCX if AMD hasn't done something interesting here, but what do I know? Also, one active core per CCX helps with heat spreading and available boost.

Furthermore, AMD producing chips with four CCX modules as an economy product is suggestive of some kind of packaging constraint. Like have a TSV substrate that's designed to straddle the corner of each of four CCX modules and not wanting—for now, while defect rates remain adequate—to produce a special tandem TSV substrate, and tool up for its assembly. Perhaps the game plan is to drive down the price of the hexacore modules until these become standard at the low end, and not bother ramping on a twin CCX assembly process for volume production at all (maybe just niche mobile products, and maybe then with a somewhat thinned CCX module a year down the road; or—a slightly bulked low-voltage CCX module with four cores and no substrate interface at all).

But I'm just using logic here, rather than digging around for revealed information, and that has only a modest track record.

The time for digging is after Naples.

Re:Crippled Ryzen 7

[Lehk228]

nah you'll enable a broken core and get a bricked CPU

Re:Crippled Ryzen 7

[Anonymous Coward]

Check out articles on Windows Ryzen Thread Bug - https://www.google.com/search?q=windows+ryzen+thread+bug&oq=windows+ryzen+thread+bug&aqs=chrome..69i57j69i60l3.4185j0j7&sourceid=chrome&ie=UTF-8, let's wait till Microsoft fixes that before we do a full comparison.

Re:Crippled Ryzen 7

[bongey]

Zen has completely new way of loading from memory, but don't take my words for it, take Linus Torvalds. http://www.realworldtech.com/f...
FUD about CCX latency most likely coming from Intel. AMd Zen ends up producing less cache misses.

Microsoft disables Windows on AMD Ryzen processors

[najajomo]

"When you try to scan or download updates through Windows Update, you receive the following error message: Unsupported Hardware .. Your PC uses a processor that isn’t supported on this version of Windows and you won’t receive updates."

Re:Ryzen = A Flop. Not Megaflops.

[Highdude702]

Well seeing as NO CURRENT CPU that is being released is going to have support for windows 7, your comment is null and void. Also, Anybody seriously considering the Ryzen 7 in a professional setting, Is going to be smart enough to disable all of that shit, Or spend the money on the enterprise versions without all the bullshit.. But most likely they will run a Unix based system on the processor and not give two fucks about what Microsoft OS they can run on the chip.

Cant decide

[Anonymous Coward]

If it is a good or a bad move to not release a higher clocket 4 core cpu. If they had it they would compete better for gaming in todays market.
But then if they do not release it they migth drive the markets to more cores and software that uses it better.

But from a technical standpoint it should be easier to reach higher clocks with fewer cores.

Re:Cant decide

[Anonymous Coward]

From a technical standpoint a fixed design on a fixed process has a fixed optimum operating frequency.
For Zeppelin that appears to be around 3.2-3.5GHz, as that's the where the V/F curve slope suddenly changes and perf/W starts to take a nose dive.
Combine that with Naples existing and it seems pretty obvious that Zeppelin was aimed to be a 30-40W die to get a reasonable 120-160W TDP on 4-die MCM Naples from the very start.

Re:Ryzen = A Flop. Not Megaflops.

[Highdude702]

Myself and other power users will stick to Windows 7 and the Intel i7 core.

Just don't let the door hit your bum on the way out, AMD. You deserve to fail, you fucking idiots.

Also sounds like you and your friends like being raped for a very small increase in IPC. Yet AMD is the one you're calling the fucking idiots.

Re:Ryzen = A Flop. Not Megaflops.

[jawtheshark]

Do you really think they are in a position to be picky? You can bet that Microsoft is behind this, in some way. Probably like "how would you like your Windows 10 drivers be delayed in certification if you produce Windows 7 drivers?". AMD needs the Windows 10 for the OEMs to even consider the chips, because -like it or not- Windows 10 is here to stay. The OEMs might want to produce Windows 7 machines, but Microsoft is going to bully them as much as they can. Look at Vista or Windows 8. Even if sales were bad with those operating systems, OEMs had to deliver them. Downgrading was just for select business machines.

Regardless... It is not clear whether those chips won't work at all or just will not deliver all functionality (power management, automatic overclocking, etc...). Newer Intel chips also are only supported on Windows 10, but they're still x86-64 chips, so it should run x86-64 code. I doubt Windows 7 will plainly refuse to run on any of these chips.

Windows 7 is EOL in three years. While I personally think it's one of the best systems made by Microsoft (and I'm a full time Linux user), it's doomed, just like XP was doomed. (Oh, and Vista is EOL next month.... Nobody is sad to see that bastard die, except of course for those people who will now be forced to buy a new machine. Like my neighbours: their machine did what it needed to do, but I expect them to come ring at my door somewhere during April.)

Re:Frankly this is what they should have launched

[Anonymous Coward]

Latency? What a crock of shit.

Re:Ryzen = A Flop. Not Megaflops.

[hairyfeet]

Besides what kind of "support" do you really need with a CPU? People have run WinXP on an AMD 9590 (it ran hellafast, surprise surprise) and had no issues so as long as the motherboard manufacturer has Win 7 drivers? I don't see the issue.

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

"When you try to scan or download updates through Windows Update, you receive the following error message: Unsupported Hardware .. Your PC uses a processor that isn’t supported on this version of Windows and you won’t receive updates."

Use Linux. I don't understand why people use OSes that dictate how they can use their computers.

Re:Microsoft disables Windows on AMD Ryzen process

[jawtheshark]

On the other hand, Windows 7 automatic update has become a clusterfuck any way. So many machines aren't getting updates any more, because one core is pegged by wuauserv.exe. Granted, it's much less likely on bare metal installations, but I have seen it. On single or dual core Virtual Machines, it's neigh impossible to get them fully updated. Especially, when they are low usage VMs just spinned up occasionally for small tasks. I just turned Windows Update off on those.

... but I doubt Microsoft is going to change that anytime soon... or ever.

Re:Ryzen = A Flop. Not Megaflops.

[jawtheshark]

as long as the motherboard manufacturer has Win 7 drivers

Often the generic stuff works just fine. In the case of Ryzen on 7 (or XP), I'd just expect to see a few warnings in the device manager. Sure, some stuff might not work (integrated USB 3.x controllers, and stuff like that)... Obviously I'd need to try, but I doubt it won't "work at all".

Re: Frankly this is what they should have launched

[Anonymous Coward]

Now you practically described my upcoming config. Build processes have plenty of stupid code that attempts to bruteforce NP-hard problems or does otherwise totally redundant stuff but you can't shortcut it because then it would become nondeterministic and you'd pay that in debugging heisenbugs.

But more cores help. 16 pvrtextools finish faster than two.

Re:Frankly this is what they should have launched

[Anonymous Coward]

They are too expensive, a cheaper Intel i5 or i3 is the way to go.

Then why are you even mentioning i3s when there's the G4560?

Re:Microsoft disables Windows on AMD Ryzen process

[hairyfeet]

Pssst....WSUS Offline or Autopatcher and Bob's your uncle...oh and you're welcome ;-)

Re:Ryzen = A Flop. Not Megaflops.

[CeasedCaring]

AMD has refused to support Ryzen on Windows 7. Instead AMD said they will only support it on Windows 10. This is stupid.

Maybe MS have convinced* AMD to make this decision?


* - With a nice, fat cheque

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

It is not disabling if you they havent implimented it yet.
Cus you are not running an OS at EOL that was made before all was known about ryzen are you?

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

For fresh Win7 x64:

* Set Windows Update to manual.
* Stop Windows Update Service and wusa.exe process if running.
* Install KB976932 (Service Pack 1),
* Reboot if necessary.
* Stop Windows Update Service and wusa.exe process if running.
* Install KB3020369 (April 2015 servicing stack update for Windows 7 and Windows Server 2008 R2)
* Reboot if necessary.
* Stop Windows Update Service and wusa.exe process if running.
* Install KB3125574 (Convenience rollup update for Windows 7 SP1 and Windows Server 2008 R2 SP1)
* Reboot if necessary.
* Stop Windows Update Service and wusa.exe process if running.
* Install KB3172605 (July 2016 update rollup for Windows 7 SP1 and Windows Server 2008 R2 SP1) or find the latest rollup and install that instead.
* Watch as Windows 7 Update actually works.
* Delete the 20 or so backported spyware updates (or hide them before they install).
* Install every new "Quality" update with paranoia at what Microsoft is sneaking in with this one.

Re:Frankly this is what they should have launched

[Anonymous Coward]

"For audio production Ryzen seems to have higher latency."

Fucking source, please, because all that shit should be handled by the (typically on-board) sound card.

"Video encoding is the one area they seem to really win at. There the more core equal more performance and you can get 8 for the price Intel sells you 6. So if that's what you are after, then it is a good deal. Not really the most common use."

Do you even fucking Twitch/Youtube/Facebook Live/Ustream/Livestream? Guess how many users there are in total on those platforms, and then the ones I haven't even mentioned, either!

"For your average user, they are way overkill."

Not with all the pre-loaded bullshit and spying, you need all the power you can get to run that shit while keeping user experience satisfactory so they don't suspect something is wrong.

Re:Ryzen = A Flop. Not Megaflops.

[Anonymous Coward]

You're a fucking moron. The processor will run Windows 7 just fine. The OS simply can't take advantage of the newer processor features present in Ryzen, which is not an AMD problem, but a Microsoft one. All the older shit Ryzen was based on ALREADY FUCKING WORKS WITH WINDOWS 7 AND THAT WON'T CHANGE, and Ryzen has all of that built into it.

In other words, there's no fucking need for AMD to support Windows 7 with Ryzen because it is essentially ALREADY SUPPORTED TO THE MAXIMUM CAPABILITY MICROSOFT WANTED.

And I bet the second I get that new Ryzen in, and install Windows 7, it will Just. Fucking. Work. Much like when I threw Windows 2000 on an FX-8350, it just fucking worked, despite the processor having zero 'support' which you're so fucking clueless about.

Re:Microsoft disables Windows on AMD Ryzen process

[jawtheshark]

Interestingly, I have two virtual machines where I did exactly that (This is documented on a few Windows fora, but Windows fora are so low in quality compared to Linux fora that they are very frustrating). Still ended up with a wuaserv.exe hogging a CPU. A Win7 without update is fine, in most use-cases for virtual machines.

Re: Frankly this is what they should have launched

[K. S. Kyosuke]

Sometimes you can't avoid solving hard problems. Although I suspect that caching solutions to them is being underutilized right now.

Re: Frankly this is what they should have launched

[Anonymous Coward]

Well, except for raidshit. M.2. 960 has more perf than the sataports combined ;)

Oh no will he make it

[future assassin]

https://scontent-sea1-1.xx.fbc...

Re:Frankly this is what they should have launched

[Anonymous Coward]

Have you even heard of VSTs? Of course CPU load IS the limiting factor, if you're using VSTs, which most people who use their computers for music are...

Re:Microsoft disables Windows on AMD Ryzen process

[armanox]

Umm...Windows 7 and Windows 8.1 aren't EOL my friend.

Pussy Footing

[Anonymous Coward]

When will CPU makers stop pussy footing around. Ever since the demise of the clock wars we have not really found a good way to improve speeds apart from multi-coring but 4 core, really!! Low TDP really! Would you want a stonking fast processor with a high TDP or a compromise We should be looking at CPU's with the following by now 16 core+ Min 4.2Ghz, maybe more with better cooling systems so we can get 5Ghz+ without resorting to liquid nitrogen

Re:Pussy Footing

[Anonymous Coward]

Your parents are paying your electricity bill eh?

Re:Frankly this is what they should have launched

[drinkypoo]

Programmers are going to be very happy with affordable 8-core chips, though. I know I will.

But programmers have already been happy with AMD's affordable 8-core chips. I have one in my PC right now (an 8350, gently overclocked) and the price:performance ratio was top-notch. It's lovely for running multiple VMs at once, too.

Re:Ryzen = A Flop. Not Megaflops.

[drinkypoo]

Also sounds like you and your friends like being raped for a very small increase in IPC. Yet AMD is the one you're calling the fucking idiots.

It very much depends on the use case. Intel processors with four cores deliver lower minimum frame rates in games than AMD processors with eight. I have an AMD processor anyway because the system came out around $300 cheaper with similar maximum performance as compared to using an intel chip, but I can notice what happens to the chip when it gets heavily loaded.

Ryzen problems with Windows 7

[Anonymous Coward]

"AMD did test Ryzen on Windows 7 where it presumably booted up just fine, but there might have been some features that didn't work as anticipated. This doesn't necessarily mean you can build a Windows 7 system around Ryzen. It should still boot and run, it just might not perform as well. There could also be bugs that cause wonky behavior, and if so, no patch will come from AMD. So, it's a roll of the dice."

AMD confirms there will be no Ryzen drivers for Windows 7 www.pcgamer.com/amd-confirms-there-will-be-no-ryzen-drivers-for-windows-7/ Feb 10, 2017 - AMD has a message for anyone clinging to Windows 7: 'No Ryzen drivers for you!'

It's official: AMD Ryzen won't support Windows 7 - Windows Report - windowsreport.com/amd-ryzen-windows-7/ Feb 15, 2017 - If you are adamant on clinging onto Windows 7, chances are you will miss out on upcoming Ryzen driver support from AMD after the Sunnyvale ...

AMD: Sorry, there will be no official Ryzen drivers for Windows 7 ... www.pcworld.com/.../amd-sorry-there-will-be-no-official-ryzen-drivers-for-windows... Feb 8, 2017 - AMD won't be providing Windows 7 drivers for its upcoming Ryzen processors, the company said, contradicting recent reports that indicated ...

Hopes dashed, AMD only supporting Windows 10 with Ryzen drivers ... https://www.pcgamesn.com/amd/ryzen-windows-7-drivers Feb 10, 2017 - Bad news everybody! Sorry, I know that's not a great way to start the morning, but if you were stubbornly sitting on a Windows 7 system and ...

AMD Denies Report Of Official Ryzen Windows 7 Drivers, Confirms ... hothardware.com/news/amd-denies-report-of-official-ryzen-windows-7-drivers Feb 9, 2017 - Well, it looks as though any hope that PC users had of receiving official Ryzen driver support in Windows 7 have officially been flushed down ...

No Windows 7 Drivers for AMD Ryzen | techPowerUp https://www.techpowerup.com/230517/no-windows-7-drivers-for-amd-ryzen Feb 9, 2017 - AMD confirmed that it will not release Windows 7 drivers for its upcoming Ryzen series processors. ...

Re:Ryzen problems with Windows 7

[Anonymous Coward]

Hmm, what do all of these have in common? Oh right, they were all written a month before release based on ... a press statement.

How about *after* release?

How To Get Ryzen Working on Windows 7 x64

Re:Ryzen problems with Windows 7

[Highdude702]

As soon as i can afford to build my Ryzen 1800X system. I will be ditching Windows as main OS and it will run from a VM on a Linux machine. And in a VM Windows doesnt give two fucks what the real processor is, Because it gets given a Generic HWID anyways. Therefor it will still Just Work! And Any optimization that is done will roll in with Linux and VM Updates.. And i will be able to reap the benefits on said Windows environment.

Re:Ryzen problems with Windows 7

[Khyber]

Processors don't need drivers. If they did, you'd be well fucked before you could ever boot, you'd be stuck bootstrapping it manually EVERY FUCKING TIME.

Re: Frankly this is what they should have launched

[Anonymous Coward]

"Intel i5 or i3 is the way to go."
Yeah sure. 2% performance difference, twice the price. Get real.

Re:Frankly this is what they should have launched

[dbIII]

I'll be real interested to see the benchmarks when they come out.

So everything above that line was based on pure guesswork?
Is this another one of those "social media workers" doing a "viral campaign" for Intel or does it just look that way?

Re:Ryzen = A Flop. Not Megaflops.

[Anonymous Coward]

Frankly "lower minimum" doesn't sound very attractive, I assume you meant higher minimum...

However, that's again putting the cart in front of the horse, since you have absolutely no idea why this happens, you're just observing a phenomena and blaming the part that is hit by it - correlation is not causation, etc. I'd take a loooong, hard look at software itself, or indeed the compiler used, before I'd start yapping about the manufacturer of the CPU.

Re:Ryzen = A Flop. Not Megaflops.

[Mashiki]

Maybe MS have convinced* AMD to make this decision?

* - With a nice, fat cheque

Only if they convinced Intel too. Because Kaby Lake isn't supported either.

Re:Ryzen = A Flop. Not Megaflops.

[drinkypoo]

Frankly "lower minimum" doesn't sound very attractive, I assume you meant higher minimum...

Er, yeah. Thanks for figuring that out, you do your species credit.

However, that's again putting the cart in front of the horse, since you have absolutely no idea why this happens,

I don't really care. It shows up in benchmarks over and over again (not just in one game or something) and it jibes with my own experiences. It was still worth it to me to go AMD, because it was so very much cheaper and if I get a few less FPS nothing bad happens, but if I have a few hundred less dollars, something bad could happen.

Re:Ryzen = A Flop. Not Megaflops.

[Billly Gates]

Google Ryzen BSOD? This happened before the review (I htink from Anatech or arstechnica) did the update to Windows 10 anniversary edition. The hardware drivers will not work with any other version including earlier versions of Windows 10.

A reviewer on youtube also put Kabylake on Windows 8.1 and it too BSOD every few minutes.

This isn't 1985 where a CPU just did arithmetic only. The CPU is the whole computer thanks to demand for tablets where Silcon on a chip is everything so an older OS today in this decade means everything including PCI express, bus, efi, Ethernet, and graphics drivers all have to be supported at the CPU as this is where the calculations are done today. Even raid is now done on the CPU. Running XP on a system is suicide as the system will not work properly due to not being optimized for specific algorithms or having drivers that function.

Boards do alot less these days too as the CPU takes over more and more functionality. In the old days before AthlonXP I remember the memory controller was board quality and chipset dependent.

Re:Lolz what a joke amd

[Megane]

Would someone tell me how this happened? We were the fucking vanguard of shaving in this country. The K7 was the CPU to own. Then the other guy came out with a three-core CPU. Were we scared? Hell, no. Because we hit back with a little thing called the Athlon. That's three cores and an aloe strip. For moisture. But you know what happened next? Shut up, I'm telling you what happened—the bastards went to four cores. Now we're standing around with our cocks in our hands, selling three cores and a GPU. Moisture or no, suddenly we're the chumps. Well, fuck it. We're going to five cores.

Sure, we could go to four cores next, like the competition. That seems like the logical thing to do. After all, three worked out pretty well, and four is the next number after three. So let's play it safe. Let's make a thicker aloe strip and call it the AthlonSuperTurbo. Why innovate when we can follow? Oh, I know why: Because we're a business, that's why!

Re:Ryzen = A Flop. Not Megaflops.

[Anonymous Coward]

No need to be so defensive, or aggressive.
My point is that "benchmarks" or specific games, or whatever you use to judge from, are pretty bloody useless tools if you want to assign "blame" since they are so insanely biased (using i387 code when running on AMD, yeah, that sounds fair...). And since I've been using AMD's exclusively since my first 486, I'm certainly not questioning your motives for choosing their products. :)

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

Because the text editors available on Windows are way better than what's available on linux. (And no I don't mean sublime, it's awful).

Re:Frankly this is what they should have launched

[KozmoStevnNaut]

Going back a little further, I've been supremely happy with my Phenom II 1100T (6-core) for years and years.

Re:Ryzen = A Flop. Not Megaflops.

[Highdude702]

As with any new chip architecture it is going to take time to iron out the bugs. The chips have been out less than a few weeks. Give devs and manufacturers some time before you talk shit about the product. let it mature a little. Its already a known fact that devs and manufacturers favor intel, so you really think they stopped everything they were working on to start throwing ryzen code around? Think about how the real world works. Things like this take time, They always have.

Re:Frankly this is what they should have launched

[drinkypoo]

Going back a little further, I've been supremely happy with my Phenom II 1100T (6-core) for years and years.

Yes, I also have a 1045T system and it is still chugging along nicely running Linux. It's got a used Asus GTS 450 OC in it, which is fine.

Re:Ryzen = A Flop. Not Megaflops.

[drinkypoo]

It's not one or two games. It's loads and loads of games. Maybe some games optimized extensively for AMD exhibit the problem less, but I haven't detected any correlation between an AMD logo splash and not having poor minimum frame rates on my AMD system. Then again, I don't have AMD graphics, because I am allergic to AMD graphics drivers.

Re:Ryzen = A Flop. Not Megaflops.

[Billly Gates]

Oh Ryzen is stable and fine with the anniversary edition of10. Just not earlier releases.

I do not understand slashdoters who get enraged and think vast conspiracy when an 8 year old OS can't run on a new CPU. Oh it must be a conspiracy by Microsoft right? The exf clocking and power are millisecond interval rocks on Ryzen dependent on UEFI and Windows 10 algorithms and drivers. No the CPU does more than process x86 instructions.

Naming convention

[Tvingo]

So the 1600 and 1700 both have 'X' counterparts but the 1400 changes to 1500X instead? Who was thinking that was a good idea?

How to use Windows 7 on new CPU (Kaby/Cannon Lake)

[Anonymous Coward]

New CPU does support Windows 7, all you need is get a USB3 expansion card to get the USB mouse/keyboard working, or use a PS/2 k/m.

For Intel graphics support, just download the latest driver, then hack the .inf file, find the latest w7_ line and replace the existing dev_id (probably a haswell dev_id, whatever the last supported gfx dev_id is), with the dev_id of your gpu (hardware id in device manager)

Once installed, play with the "Legacy Compatibility Mode" setting in BIOS, wrong settings will get you a blankscreen or bluescreen, but the right one will get you to windows normally.

Re:Frankly this is what they should have launched

[Highdude702]

Lets not pretend we didnt know this comment section was going to be filled with that shit. Some people have more money than brains. And it shows when they buy the intel CPU thats in the same class as a AMD CPU for almost double the price. Not to mention Intel's motherboards are normally another ~25% more expensive also. Even with the high end X370 boards going for almost $300. There are plenty of intel motherboards for $350+ when I walk the motherboard section at frys. And now with Ryzen the feature set is near identical. Youre just paying for that small logo that says Intel

Re:Ryzen = A Flop. Not Megaflops.

[Anonymous Coward]

One or loads and loads isn't really all that interesting. In fact since it's quite possible the problem lies deeper than that, in various libraries used or middleware - or indeed as I alluded to, the compiler used - it would be surprising if it wasn't a problem in "loads and loads" of games. E.g the Intel C-compiler is well known to outright sabotage applications by including runtime tests in the binaries produced, which switches the application to an i387 codepath instead of using MMX, SSE or whatever is available when you're running on an Intel CPU. The ICC is known as probably the best compiler you can buy, and is afaict used extensively in the benchmarking and gaming industry. And that's just one way you can tilt the results.

That means that the results come out one way, which specifically makes Intel look good, but the blame isn't necessarily with AMD, since there's very little they can do against outright cheating. Specifically, the only way to get around the problem with the ICC would be to make their CPU outright lie, and say they are "genuine Intel" processors, which would be - again, afaik, IANAL etc - illegal.

Obligatory: Intel CPU Backdoor Report

[Anonymous Coward]

Intel CPU Backdoor Report (Updated Mar 13, 2017)

The goal of this report is to make the existence of Intel CPU backdoors a common knowledge and provide information on backdoor removal.

What we know about Intel CPU backdoors so far:

TL;DR version

Your Intel CPU and Chipset is running a backdoor as we speak.

The backdoor hardware is inside the CPU/Bridge and the backdoor firmware (Intel Management Engine) is in the chipset flash memory.

30C3 Intel ME live hack:
@21m43s, keystrokes leaked from Intel ME above the OS, wireshark failed to detect packets.
[Video Link] 30C3: Persistent, Stealthy, Remote-controlled Dedicated Hardware Malware
[Quotes] Vortrag:
"DAGGER exploits Intel's Manageability Engine (ME), that executes firmware code such as Intel's Active Management Technology (iAMT), as well as its OOB network channel."

"the ME provides a perfect environment for undetectable sensitive data leakage on behalf of the attacker. Our presentation consists of three parts. The first part addresses how to find valuable data in the main memory of the host. The second part exploits the ME's OOB network channel to exfiltrate captured data to an external platform and to inject new attack code to target other interesting data structures available in the host runtime memory. The last part deals with the implementation of a covert network channel based on JitterBug."

"We have recently improved DAGGER's capabilites to include support for 64-bit operating systems and a stealthy update mechanism to download new attack code."

"To be more precise, we show how to conduct a DMA attack using Intel's Manageability Engine (ME)."

"We can permanently monitor the keyboard buffer on both operating system targets."

Backdoor removal:
The backdoor firmware can be removed by following this guide using the me_cleaner script.
Removal requires a Raspberry Pi (with GPIO pins) and a SOIC clip.

Decoding Intel backdoors:
The situation is out of control and the Libreboot/Coreboot community is looking for BIOS/Firmware experts to help with the Intel ME decoding effort.

If you are skilled in these areas, download Intel ME firmwares from this collection and have a go at them, beware Intel is using a lot of counter measures to prevent their backdoors from being decoded (explained below).

Useful links:
The Intel ME subsystem can take over your machine, can't be audited
REcon 2014 - Intel Management Engine Secrets
Untrusting the CPU (33c3)
Towards (reasonably) trustworthy x86 laptops
30C3 To Protect And Infect - The militarization of the Internet
30c3: To Protect And Infect Part 2 - Mass Surveillance Tools & Software

1. Introduction, what is Intel ME

Short version, from Intel staff:

Re: What Intel CPUs lack Intel ME secondary processor?
Amy_Intel Feb 8, 2016 9:27 AM

The Management Engine (ME) is an isolated and protected coprocessor, embedded as a non-optional part in

Re: Obligatory: Intel CPU Backdoor Report

[Anonymous Coward]

So AMD are totally immune to all that, are they?

How to use Windows 7 on new CPU (Kaby/Cannon Lake)

[Anonymous Coward]

Never give into Windows 10.

New Intel CPUs do support Windows 7, all you need is get a USB3 expansion card to get the USB mouse/keyboard working, or use a PS/2 k/m.

For Intel graphics support, just download the latest driver, then hack the .inf file, find the latest w7_ line and replace the existing dev_id (probably a haswell gpu ven_id+dev_id, whatever the last supported GPU dev_id is), with the dev_id of your GPU (check hardware id in your display adapter in device manager)

Once installed, play with the "Legacy Compatibility Mode" setting in BIOS, wrong settings will get you a blankscreen or bluescreen, but the right one will get you to windows normally.

Re:Ryzen = A Flop. Not Megaflops.

[Anonymous Coward]

I'm pretty sure power users will stick to linux.

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

Even WSUS Offline will sit there using 100% of a core for an entire day as it inefficiently tries to compute each branch of the update tree it can take.

You need to install the servicing stack update and then use Windows 7 update rollups. A short tutorial with links is located here https://www.reddit.com/r/windows/comments/4tx4s9/windows_7_slowstuck_checking_for_updates_fix_as/?st=ixfm9f3b&sh=6829197f

Re:Microsoft disables Windows on AMD Ryzen process

[ArchieBunker]

Almost daily I see svchost.exe eating up a core and thrashing my drive.

Rollups = NSA telemetry

[Anonymous Coward]

Fuck off with the rollups, we don't want fucking telemetry

Re:Rollups = NSA telemetry

[Anonymous Coward]

If you don't want telemetry, check out https://www.openbsd.org/ and put https://www.pfsense.org/ on your router.

If you want to use Windows 7, check out https://gist.github.com/xvitaly/eafa75ed2cb79b3bd4e9

If you're one of the retards who wants to sit there saying "Windows is NSAware" without even trying to contribute anything meaningful to the discussion, then carry on. We know it's NSAware. We're doing everything we can to fight it.

Re:Frankly this is what they should have launched

[Anonymous Coward]

From many of the tests I've seen the Ryzens only appear to fall behind Intel at 1080p gaming. Resolutions of 1440p or higher are on par with Intel. Also some games perform better with Intel, and some others perform better with AMD. I'm sticking with AMD, I'll just upgrade my monitor from 1080p to 1440p. I'll get better looking games, and be on par with Intel for performance.

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

I wonder how much they charged Intel for that.

Re: Frankly this is what they should have launched

[Anonymous Coward]

But those little shits run hot

Re:Frankly this is what they should have launched

[KozmoStevnNaut]

GTX 460 here, though I am considering an upgrade so I can play some newer games, or whether I should just buy a whole new system while I'm at it.

Re:Microsoft disables Windows on AMD Ryzen process

[Wdomburg]

Windows 7 hit end of mainstream support over two years ago, in 2015. Windows 8 will still be in mainstream support until January, but Microsoft announced they would not be supporting next generation hardware on the upcoming architectures over a year ago, largely due to driver support issues.

I'm sure that fueling adoption of Windows 10 is part of their motivation for the policy, but it is likely that the chip makers were not particularly interested in supporting one platform that has less than three years of extended support left and another that is nearing extended support and never achieved broad uptake in the first place.

Anyone benchmarked twitch streaming?

[rsilvergun]

E.g. does a Ryzen get better game framerates while streaming? I'm guessing not, but I've only seen raw benchmarks in the articles I've read, not "We benchmarked this game while streaming it".

Right now Ryzen doesn't make a lot of sense unless you're a programmer or video editor (either hobby or pro). If you're just a gamer it's getting beat by cheaper Intel hardware and AMD motherboards (that aren't garbage) tend to cost more pushing the price higher (though that might not be true for Ryzen). Now, given that a lot of gamers like to stream I could see that being a thing.

Re:Anyone benchmarked twitch streaming?

[TsuruchiBrian]

On paper, the Ryzen 1400 looks better than the I5-7500, and at $170 it's $30 cheaper. You can dump the money you save getting a cheaper CPU (intel or amd) into a faster GPU.

I actually do way more programming than gaming on my PC, but honestly I don't really care about compilations taking a bit longer. I am usually coding during long compilations anyway. But lag during gaming is really annoying. So I would definitely prioritize gaming performance over other sorts of benchmarks. The only other consideration I have is cost. There is so much diminishing returns in getting an expensive CPU. Is 2x the price worth 5% better performance?

I think it makes much more sense to build moderately priced desktops that cost half the price of high end desktops, and just build them twice as frequently.

Re:Microsoft disables Windows on AMD Ryzen process

[thegarbz]

Microsoft doesn't provide updates to an unsupported outdated OS on AMD Ryzen processors. ... And the latest Intel processors ... And in a few months EVERY processor since it's EOL.

Re:Microsoft disables Windows on AMD Ryzen process

[thegarbz]

Windows 7 is. It ended mainstream support 2 years ago.
Windows 8.1 isn't EOL. It was however DOA.

Re:Microsoft disables Windows on AMD Ryzen process

[Anonymous Coward]

I am using the former. Sadly, it still takes several reboots to get all the updates done (pre-IE11 updates, reboot, IE11 which is now mandatory, reboot, WU update, reboot, 117 updates, reboot), all while praying that TrustedInstaller.exe doesn't hog all the memory.

Score:0, Informative

[Anonymous Coward]

CIA/NSA shills are busy hiding this information

What I really want

[slashmydots]

What I'm waiting for is a dual core that absolutely crushes even an overclocked Pentium anniversary edition. I got mine to 4GHz off of 3.2GHz and its single thread rating beats the $1000+ i7 extreme edition from the 2011 v3 socket. Considering tabs in firefox, probably the javascript engine, most programs, and most windows tasks are single threaded, a really "snappy" and responsive PC for just facebook and web surfing would benefit greatly from a chip like that. Like a 3000+ passmark rated single thread rating on an under 75 watt dual would be veeeery impressive and very useful. As for a $169 processor for a basic use PC, no thanks. Unfortunately $400-500 PCs is about 75% of our custom orders at my shop so they're missing a huge portion of the market.

Re:What I really want

[Anonymous Coward]

What I'm waiting for is a dual core that absolutely crushes even an overclocked Pentium anniversary edition. I got mine to 4GHz off of 3.2GHz and its single thread rating beats the $1000+ i7 extreme edition from the 2011 v3 socket.

I think I've got the same chip or something close. It works fairly well on linux, and even the occasional VM. The real problem is that not all programs are well behaved. When firefox, or the OS GUI components, or the VM gets into a state which starts pegging something you start becoming hard to use fast. I bought the dual core in the theory that I wouldn't really need more than 2 and it would save power. I suppose it works most of the time, but at least for the times with a VM running, I kinda wish I'd have stayed 4+, and even without the VM, I'd almost rather have the spare resources to throw at something being buggy so it doesn't annoy me quite so soon.

On another note, I'm even slightly looking at the cheaper pcie ssd solutions again, for much the same reason. Sure you don't usually need it, but if your system gets in a bad state, or just tries to thrash, it would be nice if it, in general, kept staying responsive. Is that throwing hardware at bad software? Yeah, but I'm getting to care less.

On a final note, I have a centos server with raid 6 and file storage. You basically want to let it keep going to low power mode with the drives idle to save power, but every time you hit them you have to wait. It is tempting, to just put a single big drive in a main pc that never goes fully into sleep mode. You still have to back it up to a big external drive, and ideally two, but you completely eliminate a file server at the expense of possibly losing data that hasn't been backed up which isn't that big of a deal these days. In other words, if I went back to a 4+ core CPU machine, it wouldn't really matter if it did some file server work on the side, since most likely i'm not even at that machine when it is doing it. Also if you disconnect it before you reinstall linux, then, presumably if you did a standard format, or a standard format plus luks, getting that drive back online is easy.

Actually, I suppose you could extend the previous. Assuming you had at least two real PCs in a house, you could make them similar and then have some rsync mirroring going on or similar. That way you really only need one offsite backup. Both PCs could probably sleep, provided you had at least some time they were both awake to keep things from drifting too far out of sync. The systems would also be nearly self contained, such that any network interruption only matters for the internet. The single offsite backup would just be the hail mary protection against your house getting leveled or robbed.

Re:Ryzen = A Flop. Not Megaflops.

[gweihir]

Oh, it is not AMD (or Intel) screwing up here, it is MS, but both AMD and Intel going along with it. No Win7 support (and the latest MS Win7 rollup-patch actively refuses to install on the new CPUs) means I will cancel the about $1500 I was planning to spend on a new Ryzen system and take a vacation for the money instead. There is no way I am going to switch to the Win10 Malware in its current state. Ads, spying and impossibility to block updates are each an absolute no-go.

Re:Ryzen = A Flop. Not Megaflops.

[gweihir]

Security patches. MS has just started to make it impossible to install them on Win7 systems with the new CPUs.

Re:Microsoft disables Windows on AMD Ryzen process

[unixisc]

Uh, most people are. They'd either use Notepad, or if they are stuck in DOS, Edit. Here, on my TrueOS box, I use Lumina text editor, or whatever default editor comes w/ the DE I'm using. But I'd have to have one of those O'Reilly books handy if I had to work in either vim or emacs (let alone vi)

Re:Microsoft disables Windows on AMD Ryzen process

[erapert]

And no I don't mean sublime, it's awful

ARE YOU SERIOUS?!
But ok, let me take a breath and calm down... have you tried Gedit? Kwrite? Vi/Vim? Emacs? Atom? Bluefish? Geany? Anjuta? Leafpad? and on and on?

Re:Ryzen = A Flop. Not Megaflops.

[Narcocide]

That almost sounds like sound logic, but... why punish AMD and Linux for something that you squarely admit is Microsoft's fault?

Re:Frankly this is what they should have launched

[dbIII]

The thing that has initially impressed me the most is seeing fairly cheap desktop boards available for AMD Ryzen that can take 64GB of memory. I've had to go for more expensive SuperMicro stuff to get that in the past for AMD or Intel.
As for performance, all I can do is read reviews for now.

Of course what I really want to see is their next generation of fast multi-way CPUs to get 128 fast cores on one motherboard but that's going to have to wait.

Re:Frankly this is what they should have launched

[K. S. Kyosuke]

So the 1700 is continuing a great tradition? I'd agree.

Re:Frankly this is what they should have launched

[nctritech]

My AMD FX-9590 system is whispering to me that you're posting a few years late ;-)

Re:Ryzen = A Flop. Not Megaflops.

[nctritech]

512-bit AVX requires OS support: "The extended registers, SIMD width bit, and opmask registers of AVX-512 are mandatory and all require support from the OS."

Game over Intel in server power consumption

[bongey]

It's Game over for Intel in the server market on power consumption. A slightly under-clocked Ryzen 1700 scores 850 in cb, and draws only 30 Watts at full load. Intel's low power offering Atom c2000 CPU's draw 33-35W under full load. Intel really will loose 15-30% new server chips sales on this alone, Xeons are 90-140w under full load.https://forums.anandtech.com/threads/ryzen-strictly-technical.2500572/

FUD about CCX latency is FUD. Zen does memory access completely differently, but don't take my words, take Linus Torvalds. http://www.realworldtech.com/f...

So basically Zen ends up produces less cache misses, so just measuring pure cache latency is invalid benchmark.

That only works if my mobo isn't $30 more

[rsilvergun]

but to be fair I haven't priced out Ryzen boards. I remember that at least on newegg the 8350's price advantage largely evaporated because it really needs a nicer mobo and faster ram to be competitive with the equivalent i5.

Also, that i5 has been out longer and is better optimized, supported & understood. The $30 bucks might be worth it. Now, if it was $50, $75... $100. You'd be talking. That's a pretty big GPU upgrade and it's what I got years ago with my Athlon 64. But $30? That takes me from one 6gb gtx 1060 to a slightly faster GTX 1060...

Re:That only works if my mobo isn't $30 more

[TsuruchiBrian]

I don't think $30 makes that much of a difference. I was more talking about the savings from getting a cheaper (e.g. sub $200) CPU as opposed to getting a high end one like an i7. At the high end is where it makes sense to still buy an Intel if what you care about is gaming. In the middle it seems like AMD might be a better value unless Intel drops the price of the I5-7600K to $170 or something.

Also I haven't really looke dinto MOBOs too much. But my brief research lead me to believe that AMD MBOs were generally cheaper, leading to an even bigger price advantage. I have not heard anything about the AMD MOBOs being the source of any performance bottlenecks.

I was figuring that the total savings from going from an Intel I7 + MOBO to a middle end Ryzen + MOBO would be ~$200 which will definitely get you a better tier of GPU.

I'm sure I've spent money inefficiently on gaming computers in the past, and I am going to try really hard to do it more efficiently this time around, especially since I will probably be building 3 or 4 identical ones.

Re:Frankly this is what they should have launched

[Highdude702]

Only a few more months and there should be a Multi Socket Naples motherboard put on the market just before the server chips hit market..

Re:Microsoft disables Windows on AMD Ryzen process

[jawtheshark]

Task Manager -> Right click on the offending "svchost.exe" -> Select "Show Services"

(This is from memory, so, might vary a bit)

It now switches to the processes tab, and all services associated with that svchost.exe will be highlighted. You can bet that "wuasrv.exe" (Windows Update Service) will be amongst the ones selected.

Another way to see whether it's Windows Update, is go to the services control panel and stop the Windows Update service. If the CPU usage goes to normal, your Windows Update is messed up. I have given up trying to fix it, and just set the Windows Update service to "disabled" now.

My main OS is Linux any way, so for the really occasional use of Windows, I can live with an unpatched version. This is -of course- unacceptable for people who use it as a main OS.

Re:Frankly this is what they should have launched

[Shirley Marquez]

Good point. Even those under-$100 B350 motherboards will take 64GB; all the ones I have seen have four sockets. I have seen pictures of A320 motherboards with two sockets as well as with four, but no A320 boards seem to be available in the real world yet. The small form factor A300 and X300 boards will be limited to two sockets and 32GB when they finally appear.

I doubt that AMD designed in any artificial roadblocks; it's not their style. So that ceiling will increase when larger DDR4 sticks become available. Right now 32GB sticks are only available as registered ECC modules and consumer motherboards don't accept those, but larger unbuffered sticks should be available in the future when higher capacity SDRAM chips are released.

The birth of OC

[DarthVain]

This is what essentially birthed the whole OC enthusiast market, As when the lower clocked model became more popular and sold more they couldn't keep up with demand, and the chips all basically costing them they same, they simply took perfectly good chips that could pass the higher test, but down clocked them anyway for sale. People found out, and starting Over clocking them. I recall people trying to get chips with certain serial numbers so you could tell what batch they came from and what plant they were made in as a tell as to if you had a better chance at a "good" one or not.

Over the years this seems much less now, and pretty much everything you get is going to more less be about the same. Indeed the last CPU I bought was a "K" variant that was left purposefully unlocked for that express purpose. Though I must say it kind of takes a bit of the shine out of it...

I think I actually still have about 6 or 8 of the 66 and 100 Mhz Celerons that I bought as a lot for a dual cpu (back when that was two physical processors) rig I made on an Abit BP6 (with pretty golden orbs coolers)... I believe it dies how it lived, playing Warcraft 2 or 3 or possibly Enemy Territory... damn that was a great game...