AMD's Future Chips & SoC's: News, Info & Rumours.

[juanrga]

Summary of Ryzen 2000 series reviews.

https://www.3dcenter.org/news/ryzen-2000-launchreviews-die-testresultate-zur-anwendungs-performance-im-ueberblick

The site has compiled most of the Ryzen 2000 series reviews, so far 26 reviews present.
The interesting part is the average of those reviews: 2700X is 8% faster than 8700K while 2600X is 13% faster than 8600k.

So the 2700K is 8% faster despite having 33% moar cores. And the 2600K is 13% faster despite having 2x moar threads.

 

[-Fran-]

Summary of Ryzen 2000 series reviews.

https://www.3dcenter.org/news/ryzen-2000-launchreviews-die-testresultate-zur-anwendungs-performance-im-ueberblick

The site has compiled most of the Ryzen 2000 series reviews, so far 26 reviews present.
The interesting part is the average of those reviews: 2700X is 8% faster than 8700K while 2600X is 13% faster than 8600k.

So the 2700K is 8% faster despite having 33% moar cores. And the 2600K is 13% faster despite having 2x moar threads.

Turbo speeds are also lower for the 2700K, don't forget that "detail". 4.7Ghz vs 4.3Ghz in single threaded tests is a ~10% speed difference on Intel's favor. And, assuming that Intel's turbo speeds can hold higher clocks, you can also say that the 10% difference in clocks stays across core loading.

There was a graph that showed how the 8700K manages the clocks with good cooling... It was a few pages back when the 8700B was announced.

Also, why does it even matter? XD

"It don't matter if you win by an inch or a mile. Winning's winning.", Toretto. Harhar!

 

[aldaia]

Summary of Ryzen 2000 series reviews.

https://www.3dcenter.org/news/ryzen-2000-launchreviews-die-testresultate-zur-anwendungs-performance-im-ueberblick

The site has compiled most of the Ryzen 2000 series reviews, so far 26 reviews present.
The interesting part is the average of those reviews: 2700X is 8% faster than 8700K while 2600X is 13% faster than 8600k.

So the 2700K is 8% faster despite having 33% moar cores. And the 2600K is 13% faster despite having 2x moar threads.

So they offer you a FASTER and CHEAPER car, and you complain that it has 6 cylinders instead of 4.

 

[Rogue Leader]

Just going to drop this here.

TDP != Power Consumption.

TDP = sustained power consumption

A processor with a TDP of 140W should, in general, consume 140W of power (plus some efficiency losses).

EVERYONE

Let me make one thing extremely clear here, based on the past page of conversation. Stop with the calling people out.
We have discussed this.

However i want to make something else very clear. A lot of bad information has been getting passed in here. This information has instigated much of the arguments and unrest. The next person who posts something with selective editing or anything else, which clearly intends to cause unrest, is GONE. Using the above link as an example:

Second, what is the sustained power consumption? This metric is widely called the thermal design power (TDP) because it determines the cooling requirement. TDP is neither peak power, which is often 1.5 times higher, nor is it the actual average power that will be consumed during a given computation, which is likely to be lower still. A typical power supply for a system is typically sized to exceed the TDP, and a cooling system is usually designed to match or exceed a TDP.

This is an extremely clear example. And it will not be tolerated, period.

 

[goldstone77]

A Timely Discovery: Examining Our AMD 2nd Gen Ryzen Results
by Ian Cutress & Ryan Smith on April 25, 2018 11:15 AM EST
Conclusion: It Changes Our Results

https://www.anandtech.com/show/12678/a-timely-discovery-examining-amd-2nd-gen-ryzen-results

 

[-Fran-]

Oh, beat me to it, goldstone77.

One hell of a followup by Ian. The information he got from investigating his results are so interesting. In fact, it might even explain why Intel has issues with stuttering and streaming!

EDIT: https://www.phoronix.com/scan.php?page=article&item=ryzen-2700x-bios

BIOS'es with the AGESA upgrade might bring some nice improvements to RyZen2 across the board, with a few regressions. They average a positive gain.

Cheers!

 

[jdwii]

Summary of Ryzen 2000 series reviews.

https://www.3dcenter.org/news/ryzen-2000-launchreviews-die-testresultate-zur-anwendungs-performance-im-ueberblick

The site has compiled most of the Ryzen 2000 series reviews, so far 26 reviews present.
The interesting part is the average of those reviews: 2700X is 8% faster than 8700K while 2600X is 13% faster than 8600k.

So the 2700K is 8% faster despite having 33% moar cores. And the 2600K is 13% faster despite having 2x moar threads.

Turbo speeds are also lower for the 2700K, don't forget that "detail". 4.7Ghz vs 4.3Ghz in single threaded tests is a ~10% speed difference on Intel's favor. And, assuming that Intel's turbo speeds can hold higher clocks, you can also say that the 10% difference in clocks stays across core loading.

There was a graph that showed how the 8700K manages the clocks with good cooling... It was a few pages back when the 8700B was announced.

Also, why does it even matter? XD

"It don't matter if you win by an inch or a mile. Winning's winning.", Toretto. Harhar!

Really just like we all knew the reason why Ryzen was under performing in games was always over the latency and lower clock rate and Amd improved this making them closer in games was actually a nice surprise

 

[jdwii]

Oh, beat me to it, goldstone77.

One hell of a followup by Ian. The information he got from investigating his results are so interesting. In fact, it might even explain why Intel has issues with stuttering and streaming!

EDIT: https://www.phoronix.com/scan.php?page=article&item=ryzen-2700x-bios

BIOS'es with the AGESA upgrade might bring some nice improvements to RyZen2 across the board, with a few regressions. They average a positive gain.

Cheers!

One other thing that is happening is some motherboard manufactures are having issues with lower bandwidth for example on the current 4.6 bios of the asrock taichi and members at overclocker are hoping its resolved.

This time around is a lot better then when Ryzen first launched though haha

 

[randomizer]

This time around is a lot better then when Ryzen first launched though haha

I'd be very worried if it wasn't! Though I have to say, the fact that BIOS updates are still making substantial differences to performance beyond edge cases is still concerning. This isn't a brand new platform anymore.

 

[-Fran-]

https://techreport.com/news/33562/amd-enjoys-massive-revenue-and-profit-growth-in-q1-2018

Looks like AMD is indeed "Ryzen" from the reds. Sorry, I couldn't help it.

Cheers!

 

[goldstone77]

Time Stamps are below

Edit:
http://ir.amd.com/financial-information/quarterly-results
https://edge.media-server.com/m6/p/6fina885
19:23 question and answers starts
2nd edit:32:40 sampling 7nm Vega and 7nm Epyc this year
3rd edit: 44:20 7nm Zen 2 will sample later this year to customers, and that will be in production 2019.
4th edit: 54:30 12nm is performance improvements it is not a die shrink
5th edit: 39:57

Our foundry strategy is to use both TSMC and GlobalFoundries. On the first 7nm product we are using TSMC for that product.

I'm not concearned about compacity.

6th edit: Deleted first statement and leaving time stamps.
7th edit: Earnings call Transcripts
https://seekingalpha.com/article/4165998-advanced-micro-devices-amd-q1-2018-results-earnings-call-transcript?part=single

 

[goldstone77]

Jim Keller is going to work for Intel!
http://www.tomshardware.com/forum/id-1581001/intel-future-chips-news-rumours-reviews/page-62.html#20917873

 

[goldstone77]

4GHz CPU Battle: AMD 2nd-Gen Ryzen vs. Intel 8th-Gen Core
Instructions Per Cycle Shootout
By Steven Walton on April 26, 2018

Spoiler

Gaming

Spoiler

Power Consumption Comparison

Spoiler

Closing Remarks

Despite a rather large clock speed deficit, the second-gen Ryzen CPUs aren't often that far behind their Intel rivals in application benchmarks and here we see why when comparing them clock-for-clock at 4GHz. In applications such as Cinebench R15 we see that the single core performance is down just 3% but where SMT is well leveraged AMD can be up to 4% faster.

We found that AMD was 3% slower in the Corona benchmark but much the same for our Excel, V-Ray and video editing tests. Then while it was 15% slower in HandBrake it was also 8% faster for the PCMark 10 gaming physics test. Of course, there is still the matter of gaming and I bet a few AMD fans were hoping we could put most of the gaming performance deficit down to clock speed. Sadly though, that's not the case.

https://www.techspot.com/article/1616-4ghz-ryzen-2nd-gen-vs-core-8th-gen/

 

[-Fran-]

https://gizmodo.com/inside-amds-quest-to-build-chips-that-can-beat-intel-1824064984

Interesting read of the past, the present and potential future.

Cheers!

 

[jdwii]

4GHz CPU Battle: AMD 2nd-Gen Ryzen vs. Intel 8th-Gen Core
Instructions Per Cycle Shootout
By Steven Walton on April 26, 2018

Spoiler

Gaming

Spoiler

Power Consumption Comparison

Spoiler

Closing Remarks

Despite a rather large clock speed deficit, the second-gen Ryzen CPUs aren't often that far behind their Intel rivals in application benchmarks and here we see why when comparing them clock-for-clock at 4GHz. In applications such as Cinebench R15 we see that the single core performance is down just 3% but where SMT is well leveraged AMD can be up to 4% faster.

We found that AMD was 3% slower in the Corona benchmark but much the same for our Excel, V-Ray and video editing tests. Then while it was 15% slower in HandBrake it was also 8% faster for the PCMark 10 gaming physics test. Of course, there is still the matter of gaming and I bet a few AMD fans were hoping we could put most of the gaming performance deficit down to clock speed. Sadly though, that's not the case.

https://www.techspot.com/article/1616-4ghz-ryzen-2nd-gen-vs-core-8th-gen/

I begged so many reviewers to do this most said it would be confusing to their viewers if they did this and i said that is what reviews used to include back in the day happy to see Steve do this.

Basically 5% behind in applications per cycle and 10% behind per cycle in games. Except with AVX2 which is something Amd can fix later

 

[goldstone77]

I begged so many reviewers to do this most said it would be confusing to their viewers if they did this and i said that is what reviews used to include back in the day happy to see Steve do this.

Basically 5% behind in application per cycle and 10% behind per cycle in games except with AVX2 which is something Amd can always add later with Ryzen 2 which will have samples this year

I think like what we seen with 12nm Ryzen AMD needs to keep aggressively attacking latency. With 7nm we should see a very nice improvement in frequency, which is going to put Ryzen in a dominant position or very competitive if Intel can fix 10nm.

 

[goldstone77]

AMD confirms it's started developing 'exciting' Zen 5 architecture
But you probably won't see it until, er, 2021
[video="https://www.youtube.com/watch?v=iQ_4C2TKHQ0&feature=youtu.be"][/video]

With the video intended to tease people about AMD's upcoming releases, the executives explained that the company is always expanding its product roadmap and that the Zen 5 will be a massive release.

However, they admitted that improving the performance of new processors is not an easy task, but the Zen 5 should be worth it. Mike Clark called the developments "very exciting", adding: "As an architect, I am already working on Zen 5, actually."

Clark explained that because these processors take so long to design, he is constantly working on future generations to ensure that AMD always has a new product to put on the market, which he admitted was "kind of crazy".

AMD is working on a number of ambitious projects. By 2020, it hopes to have released Zen 2 and Zen 3, before bringing Zen 5 products to market at some point in 2021. The devices will be manufactured on sub-7nm process technology.

The news comes as, a few days ago, clues that AMD is working on the Vega20 GPU were found in Linux driver code. It is expected to launch between the third and fourth quarter of 2018. µ

https://www.theinquirer.net/inquirer/news/3029983/amd-confirms-its-started-developing-exciting-zen-5-architecture

 

[randomizer]

https://www.techspot.com/article/1616-4ghz-ryzen-2nd-gen-vs-core-8th-gen/

I wish the reviewers would mention the specific versions of the software they're using. For example, TH uses a version of Handbrake that uses builds of x264 and x265 from no later than Jan 2016. These predate both Zen and Coffee Lake and reduce the relevance of the results, but at least you know this.

 

[goldstone77]

https://www.techspot.com/article/1616-4ghz-ryzen-2nd-gen-vs-core-8th-gen/

I wish the reviewers would mention the specific versions of the software they're using. For example, TH uses a version of Handbrake that uses builds of x264 and x265 from no later than Jan 2016. These predate both Zen and Coffee Lake and reduce the relevance of the results, but at least you know this.

Yeah, lots of reviews seem to have their setting sliders moving all around. You have to really pay attention to detail!

 

[daerohn]

for AMD the real issue seems to be the fabrication. Intel has the lead here with higher clocks and ofcourse more die space. Adding AVX512 to ZEN should not be the number 1 priority. Right now they must focus on the production and increase the clock speed of Ryzen. 4.3Ghz is nothing compared to what Intel chips can achieve, with overlocking also.

further more they need more space on the die to implement AVX512. If they want to. With the current 12nm node they can not add it without increasing the die size and even if they implement it the performance gain would be limited only to some programs.

AMD is on a good path. The clock to clock performance values seem to be minor in single thread. however this still doesn't negate the truth that intel chips hit higher frequencies.

Also the news about Zen 5 and Intel new generation chips are fascinating yet I am more interested in the next gen Zen, Zen3 and next gen node. Will 7nm be better than Intel's 10nm?

 

[-Fran-]

https://techreport.com/review/33568/gaming-and-streaming-with-amd-second-gen-ryzen-cpus

More information about games and a little streaming test that reinforces other site's findings to a degree.

Cheers!

 

[Gon Freecss]

<Non English link removed by moderator>

<Please keep it in English here>

 

[jdwii]

for AMD the real issue seems to be the fabrication. Intel has the lead here with higher clocks and ofcourse more die space. Adding AVX512 to ZEN should not be the number 1 priority. Right now they must focus on the production and increase the clock speed of Ryzen. 4.3Ghz is nothing compared to what Intel chips can achieve, with overlocking also.

further more they need more space on the die to implement AVX512. If they want to. With the current 12nm node they can not add it without increasing the die size and even if they implement it the performance gain would be limited only to some programs.

AMD is on a good path. The clock to clock performance values seem to be minor in single thread. however this still doesn't negate the truth that intel chips hit higher frequencies.

Also the news about Zen 5 and Intel new generation chips are fascinating yet I am more interested in the next gen Zen, Zen3 and next gen node. Will 7nm be better than Intel's 10nm?

Well with Intel delaying their 10mn node to next year its all up for global foundries to deliver for once and release 7nm on time meaning both will be at a similar process. So really i think it's only going to get better for Amd i mean unless Intel has something hidden in the works and i'm not talking about a 8 core cannon lake i mean something with actual IPC gains

Then again this sucks
https://www.tomshardware.com/news/amd-intel-jim-keller-hire,36963.html

"Intel Hires Ex-AMD Lead Architect Jim Keller, Confirms Technology And Manufacturing Group Changes" Just shows even employee's have no kind of fanboyism why should consumers haha

 

[goldstone77]

Keller is a famous engineer, I imagine he goes wherever he can find the most exciting opportunity. Intel has an endless budget.

 

[goldstone77]

Also the news about Zen 5 and Intel new generation chips are fascinating yet I am more interested in the next gen Zen, Zen3 and next gen node. Will 7nm be better than Intel's 10nm?

I don't know about better core for core, but it will be a very close race from the looks of it. Intel will not have it's typical process advantage over AMD.

While a move from 14 nm to 7 nm was expected to provide, at the very best, a halving in the actual size of a chip manufactured in 7 nm compared to 14 nm, Gary Patton is now saying that the are should actually be reduced by up to 2.7 times the original size. To put that into perspective, AMD's 1000 series processors on the Zeppelin die and 14 nm process, which come in at 213 mm² for the full, 8-core design, could be brought down to just 80 mm² instead.

Patton said that he expects this design to be able to scale pretty well to some 5 GHz operating frequencies.

https://www.techpowerup.com/242148/globalfoundries-7-nm-to-enable-up-to-2-7x-smaller-dies-5-ghz-cpus