RYZEN Achieves 5Gz on Air (Easter Egg)

[Cryptic VI]

There is an easter egg in the Article from last week. There's a digital version of the first pages of the publication. Go to the last page in there (page 10). You'll see a long, binary string.
If you were to put that string through a binary to plain text converter...
010110100110010101101110010011110100001101000000010000010110100101110010001111010011010101000111
it says
ZenOC@Air=5G

https://www.reddit.com/r/Amd/comments/5krghq

 

[bob4933]

Geeze.... I might be jumping ship... again...

 

[Saul Goodman]

It's single core and means basically nothing. Good hype piece though that people are going crazy over.

 

[Alaric]

Seems like encouraging news, regardless. Obviously not carved in stone as an indicator, but better than finding out it caught fire at 3.8 GHz and took 5 volts to get there. LOL

 

[Woomack]

Funny is that all are asking for 5GHz but they won't see any difference in games past 4GHz. I'm not expecting AMD to be popular in workstations anytime soon because of last ~7-8 years.

 

[Alaric]

I don't know about that. Corporations (Dell, HP, etc.) don't have memories. They look at tomorrow's profit margin. If Dell pushes them, they'll sell. I doubt they'll displace Intel any time soon, but if they actually live up to the hype this time AMD should make inroads in to every market. "If" and "should" being the operative words here.

 

[Woomack]

Corporations don't trust AMD as their customers don't trust AMD. AMD is only in cheaper office and home series. I work in IT for ~13 years ( distribution, services etc ) and believe me, AMD business hardware is not selling for long years. There are almost no desktop/laptop office series and barely anyone is selling their servers. Only gamers/enthusiasts are waiting for new AMD and not even as many as we may think. Most forgot about AMD long time ago.
It's really hard to fix lost trust on the business market and office computers are maybe 60% whole computer sales. I can only tell you that I sold 1 AMD based PC in last 7 years. Not because I don't have parts in local distribution but because our clients don't want AMD.

IF they stand up and start to build up their brand from the beginning then they have a chance to appear on the market in next 3-4 years. Right now we only hear promisses and we see next delays on which AMD is only losing.

 

[Alaric]

I don't think the situation is that grim over here. I may be wrong, but it seems that way to me. You make it sound like AMD has one foot in the grave and the other on a banana peel. Honestly, your posts for a couple years have had AMD at Death's door yet they still seem to be in business and coming out with new products. I just don't see their future being as dark as you seem to. Like I said, I may be wrong. I'm certainly not touting my opinion as anything other than my opinion, for the little that's worth.

 

[Cryptic VI]

Looks like CPC published a response to it's original article. I have translated the french using google translator. Key takeaway here is "In spite of everything, we know with almost certainty that the CPU that we used for the tests actually came close to the 5 GHz with an (huge) air-dissipator. "

http://www.cpchardware.com/cpc-hardware-n31-precisions-elucubrations/

Obviously, the publication of the last issue of CPC Hardware did not go unnoticed on the Internet. If we suspected that some eddies would occur with our first exclusive tests of the AMD Zen architecture, we had not anticipated such a surge. Here are some details about the (many) questions we received directly and on social networks, both on form and substance, and also some thoughts on the media coverage of these tests.

On the diffusion of tests on the Internet

Page 61 of the magazine
At the time of the closure (early December), we had discussed a possible communication strategy around this new issue. Dandu suggested putting the famous benchmarks and a brief commentary in French and English on the day of release. For my part, I thought that a delay of 3-4 days was necessary, so that our subscribers and readers of the first hour retain exclusivity. In any case, we obviously knew that the tests would circulate quickly on the forums and that the websites would comment on them. But maybe not as fast and in this way. Just a few hours after the official release of the magazine on December 23, a user of the forum overclock.net has decided to post - of course without our agreement - a scan of the page in question. This one then spread quickly from forum in forum until making the "one" of sites like wccftech. More embarrassing to our eyes: some French sites have also reproduced in extenso the page in question, in defiance of the most basic ethical rules. If we do not see any problem with our colleagues doing their own analysis on the basis of our results, it is rather unpleasant to see his own texts copied / pasted shamelessly. The right to quote does not in any way legitimize the full resumption of an entire page. Among the journalistic pearls noted by the occasion are a colleague who says "Reddit" and another who considers that after having "broken" an NDA, it becomes "complicated to give lessons". But enough whining, let's talk about it, precisely, of the NDA.

About NDA in general
Non-Disclosure Agreements allow two partners to exchange confidential information. Concluded between two companies, the NDA represent an important tool to protect the strategy of a company. First of all, there is an obvious detail that many seem to ignore: an NDA is not a law, it is a mere trade agreement between two parties. No one is bound to comply with it if he has not signed it. Readers may think that the NDA apply to everyone from the moment a builder has decided it unilaterally shows how this system has perverted relations with the press. Formerly intended to protect industrial secrets, the NDA is now a formidable means of control which allows to include the media in the marketing plan. And the results exceed all hope.

For example, AMD did not hesitate in the past to muzzle the press with an NDA "running" several months after the sale of some of its APU. Despite this aberrant situation, the overwhelming majority of the media nonetheless respected it to the letter. More blatantly: Intel's Kaby Lake processors have been on sale in Asia since mid-November in large quantities. We have also obtained ours by this means, at the normal selling price and delivered within 48 hours by DHL. If some media have used this channel to publish timid preview, one may wonder what is holding back the bulk of the troops. The fear of violating a virtual NES and the big eyes of Intel? The absence of the unnecessary Z270 chipset can not be a valid excuse. On our side, we believe that journalists do not have to sign NDAs supposed to apply between "partners" with builders.

On that of the AMD Ryzen in particular
We were able to read on some English-language forums comments written in this way: "How damn those naughty French can have access to AMD Zen ?!". And that's a good question: why do not we and others? And the corollary: can we trust them? First and foremost, the difficulty is not having access to a Zen AMD. The Engineering Sample now circulates widely and we know several other journalists - and some overclockers - who have been able to perform tests on a functional platform. Not to mention the "secret" presentation that took place at AMD in early December in the presence of many happy few of the world press. Today the real challenge is to have access to an AMD Zen * ET * to reach agreement with the source to publish the results. For that, months (or years) of cronyism as well as a visceral tendency to paranoia are not enough: it is essential that your source has the absolute certainty that it will not be compromised in case you come Big eyes.

In most newsrooms, however, there is no shortage of brands. Among them are the call to N + x, not necessarily as jumpy as you on protecting sources. If you do not have a hierarchical superior, it remains the weapon of mass destruction: the stop of the tap to samples. Given the price of graphics cards and processors, this would entail a considerable financial burden that hardly any editorial wants to take the trouble to assume. And even if a media decides to devote a substantial budget to buying all the products tested, the delays of sale would make it very difficult to publish the tests "0-day", indispensable to survive on the Internet. Finally, last very effective lever: the pub. All websites derive the bulk of their revenue, it offers a formidable pressure - see a right of life or death - to the biggest advertisers / builders. CPC Hardware is fortunate to be unaware of any pressure of this kind, which allows our sources to trust us fully. No doubt that the past is also in our favor: some observers just recalled our preview of the Athlon 64 in 2003 on x86-secret.com, 8 months before its release. For the occasion, we offer you the non-blurred photo of the platform of the time. Rendez-vous in 2030 for that of Zen!

On the tests themselves
A point that comes up regularly in the comments concerns the performance gain noted "about 35% at equal frequency" in front of an FX-8370 (which we recall as a Quad Core). If the quote does not seem to "paste" directly with the graph, it is because it comes from the raw results and not from the average. The comparison concerns a single core with CMT or SMT in both cases (it is to say one core / 2 threads for Zen and 1 core / 2 clusters for the FX-8370). In the latter case, the definition of "core" and "cluster" corresponds to the technical definition (as explained in this AMD patent, page 6) and not in the definition invented subsequently. It would have been simpler to publish an example with raw numbers, but when negotiating with our contact, it was decided not to publish raw numbers. The reason is easily explained: these are very precise and very reproducible. AMD could therefore easily guess which combination ES / BIOS / AGESA CPU was used to perform the test ... and potentially compromise the source.

Another important point to remember, which does not appear on the scanned page that has been circulating everywhere for a week (but on the previous page): At the time of the tests, there were two types of BIOS / AGESA with quite different performances. The fastest is the one we tested. The other offers a significantly lower IPC, barely superior to that of Intel's Sandy Bridge cores. The explanation could come from one of the hardware bugs present on the first prototypes at the level of the μop cache and the SMT. We have no reason to think that these problems - quite normal at this stage of development - will not be adjusted for the output of the CPU, but we still keep in mind the fiasco of the TLB bug during the output Of the Phenom ...

ZenOC @ Air = 5G
We regularly hide some easter eggs in the magazine. In March, we had coded "Intel GPU = AMD" in a binary string of the CPU page of the purchase guide. Virtually no one had seen the information at that time when it made a big noise 6 months later. In short, we are teasing. The presence of the string of the current number, which is decoded in "ZenOC @ Air = 5G" in this issue has been chatting on the forums for 2 days. Being unmasked, we owe you some details. First, we did not summarize a test in a few bits. If we had been able to test ourselves overclocking, we would have told you openly in the preview. In spite of everything, we know with almost certainty that the CPU that we used for the tests actually came close to the 5 GHz with an (huge) air-dissipator. The I / O multiplier is not clamped at this time and is configured in steps of 0.25x. One heart, however, was active; The Motherboard VRMs seemed at that time too unstable to test with all of the cores. Other Ryzen ES are currently in the hands of overclockers and you should not delay to learn more: a demonstration of overclocking could occur at the CES if good results are achieved.

Finally, if you have not yet purchased the latest Hardware CPC, now is the time to visit your favorite kiosk: not only will it soon be collector, but it also abounds with crisp information behind the scenes of Intel ...

 

[rescuetoaster]

Corporations don't trust AMD as their customers don't trust AMD.

That might be a bit of a stretch considering both major consoles are using their hardware.

 

[ShrimpBrime]

No where is there a claim that AMD ZEN could even do 4ghz yet. Why would 5ghz on a single core be anything impressive??

 

[Mr.Scott]

No where is there a claim that AMD ZEN could even do 4ghz yet. Why would 5ghz on a single core be anything impressive??

NDA in effect.
AMD is much more serious about it than Intel is.

 

[Janus67]

That might be a bit of a stretch considering both major consoles are using their hardware.

Corporate purchase are not including those that buy them/license them for consoles. Corporations as in corporate/workstatiom/office purposes, buying from an OEM.

In my 12 years doing IT, I've never seen a company - purchased product that used an AMD cpu (desktop, laptop, server). AMD gpus were commonly bundled in our Dell (and other OEM) computers, but never an AMD cpu.

 

[Alaric]

They're selling those Opterons to somebody, aren't they??? (That's a question far more than a statement.)

 

[Cryptic VI]

Intel Kaby Lake DOA! Reviews of Intels processor showing very little performance gains. This is now the official turning point where Intel starts losing market share to AMD.

Typically with each new generation of Intel CPUs brings about a fundamental increase in performance through the rate of instructions per cycle/clock (IPC) that the processor can go through. That being said, Intel has stated (and we've confirmed through testing) that Skylake and Kaby Lake are identical for IPC. As a result, the i7-7700K attempts to take the performance crown through frequency alone. The i7-6700K runs as 4.0 GHz base and 4.2 GHz turbo, while the i7-7700K runs at 4.2 GHz base and 4.5 GHz turbo.

http://www.anandtech.com/show/10968...91w-review-the-new-stock-performance-champion

 

[bob4933]

Intel Kaby Lake DOA! Reviews of Intels processor showing very little performance gains. This is now the official turning point where Intel starts losing market share to AMD.

Typically with each new generation of Intel CPUs brings about a fundamental increase in performance through the rate of instructions per cycle/clock (IPC) that the processor can go through. That being said, Intel has stated (and we've confirmed through testing) that Skylake and Kaby Lake are identical for IPC. As a result, the i7-7700K attempts to take the performance crown through frequency alone. The i7-6700K runs as 4.0 GHz base and 4.2 GHz turbo, while the i7-7700K runs at 4.2 GHz base and 4.5 GHz turbo.

http://www.anandtech.com/show/10968...91w-review-the-new-stock-performance-champion

Much as we criticize not slamming AMD's door shut, the same should be said about Intel. There are other reasons for kaby lake. Intel has openly stated its change of policy away from "tick-tock" to process optimization. The process change is more important on the mobile computing side of things, not so much the desktop range; but you're also getting 24 pcie lanes and other stuff too.

AMD took a gamble on their FX line and lost huge. That said, their business model is actually incredibly outstanding and worth a read up on. We still have no useful data, and predictions about the market change is pointless at this juncture. When we have a 4c/8t ryzen chip that is identical to a 4790k for 150$, then yeah, Intel is gonna sweat bullets over their i7 lineup. Given the way the desktop community market has been going, it probably wont even matter what AMD say the MSRP is. The rx 480 was a "199$ card" upon release, and vendors were selling them for almost 300$ for a while. The cost will be what the market will bear; and markups will happen. Don't hold your breath on some miracle chip that gives top level performance for peanuts.

 

[bob4933]

Corporate purchase are not including those that buy them/license them for consoles. Corporations as in corporate/workstatiom/office purposes, buying from an OEM.

In my 12 years doing IT, I've never seen a company - purchased product that used an AMD cpu (desktop, laptop, server). AMD gpus were commonly bundled in our Dell (and other OEM) computers, but never an AMD cpu.

My business exclusively uses AMD, and we have thousands of workstations (company wide)... 90% of them are AMD A8/A10 cpu's. I'll find out who our OEM is, probably a japanese company tbh. There are more companies than Dell or HP out there