Discussion AMD Ryzen MegaThread! FAQ and Resources

[8350rocks]

Ok, so I took TechSpot's review, and did the math:

Out of 32 tests (16 games at 1080 and 1440), 15 resulted in improved performance with SMT disabled. A few got worse, and the rest had no effect.

But if you consider they solve the SMT issue, then only those games who had better performance without it will see gains, while the others will not (since they look already optimized, or has no effect). So, I did the difference in each scenario where SMT hurts performance, and found this (sorry for the bad table):

SMT - no SMT
min/avg - min/avg == %min/%avg (game)
56/94 - 59/96 == 5,4/2 (hitman)
50/71 - 53/72 == 6/1 (civ)
47/70 - 52/70 == 10,6/0 (civ)
133/166 - 146/170 == 9,8/2,4 (ow)
95/136 - 98/139 == 3,2/2,2 (ow)
104/126 - 106/131 == 1,9/4 (gears 4)
102/126 - 105/130 == 2,9/3,2 (gears 4)
57/79 - 58/93 == 1,8/17,7 (deus ex)
85/120 - 99/128 == 16,5/6,7 (F1 2016)
87/111 - 90/114 == 3,5/2,7 (F1 2016)
65/78 - 71/88 == 9,2/12,8 (TW Warh)
65/79 - 70/86 == 7,7/8,9 (TW Warh)
87/104 - 96/111 == 10,3/6,7 (GTA V)
60/88 - 65/100 == 8,3/13,6 (FC Primal)
61/86 - 66/93 == 8,2/8,1 (FC Primal)

which resulted in 6,98% average increase in min FPS, and 6,13% for average FPS. Highlights for F1 2016 at 1080p, for a 16.6% increase in minimum FPS, and Deus Ex at 1080p, with a 17.7% increase in average FPS.

Also, this only takes into consideration the games that won't take advantage of SMT instead of being hindered by it, in which case performance would grow even higher.

Solving the SMT problem will bring A LOT to the table. It won't solve the latency issues, but many reviews would have to be redone, and I think conclusions would change. AMD realy should have waited a bit more to launch Ryzen.

Nice work. Note that 6--7% average you got is the result of manually enabling/disabling SMT after testing which is the better option on a individual title per title basis. This is a best case scenario. No one is going to provide a scheduler with specific profiles for each one of the games that exist or will exist in a future. A real scheduler will provide a schedule on the fly less optimal. Thus a supposed future patch will provide an average performance gain of about 5% or less.

Recall what happened with the Bulldozer patches for Windows. Installing them increased performance on average by a ridiculous 2--5%. We are in a similar scenario now.

This supposed 5% extra performance from some future Ryzen patch doesn't solve the ~20% deficit that RyZen has compared to BDW-E.

Ryzen is 1% ahead of BDW-E in single thread, according to a conversation with Robert Hallock. They are 6.8% behind kaby lake, by their own admission.

Unless you think a 200-300 MHz advantage over Ryzen is going to magically produce 20% somehow when BDW-E is starting 1% behind at same clocks. I doubt a 4.1 GHz Ryzen loses by much at all to a 4.4 GHz 6900K in anything that is optimized.

The *true* performance increase will come from the April windows patch when the scheduler recognizes SMT cores. Microcode will clean up quite a bit of things, perhaps they pick up 5-10% in some areas...but the windows scheduler is what is really killing them at the moment.

 

[jaymc]

Thanks for your input Gamerk wise words I'm sure..

Here's some info I found on gaming benchmarks on Linux for the Ryzen 1800x, there few an far between and no graphics I'm afraid.

I had heard that the CPU scheduling problem is already fixed on Linux (could be wrong or maybe it's not on this distro).

Anyhow zen didn't seem to fair much better acording to this website:
http://www.phoronix.com/scan.php?page=article&item=amd-ryzen-gaming&num=1

 

[cdrkf]

To the various people replying against the vid I posted you are missing the point. This isn't a comparison to prove the fx 8350 was a better purchase. There are 3 key points that are very valid:
- it's looking at a lower core count part vs a higher one. If you want to compare to Ivy, compare to the ivy i5, the trend is the same
- it's looking specifically at the 'fact' that game reviews use to test 'cpu gaming performance' by using unrealistically lowered resolutions to 'predict' future performance.
- what it shows is very clear. The higher core count part doesn't get slower at gaming as time goes on with much more powerful gpus vs the lower core count part. In fact the delta between the two diminishes over time in favor of the higher thread count. This is despite the inherent weaknesses in fx.

How this relates to ryzen: long term it suggests ryzen should maintain pace with or possibly catch current gaming parts like the i7 7700k. Despite all the comments to the contrary higher core count parts are the better option for longevity (the same is true for Intel higher core count parts, so no this doesn't mean ryzen will catch the 6900k just like fx doesn't outpace the Ivy i7). The critical point though is in 3 years time with a gpu 3x the speed of Pascal Titan, ryzen will be in the same relative position (or arguably a tad better) vs Intel and not massively behind which is what the reviewers are trying to infer with the low resolution benchmarks. They are jumping to a conclusion that simply isn't backed up by any data.

 

[jdwii]

I rarely do this but i will say that once again something seems off and maybe we all need to wait

http://www.guru3d.com/articles-pages/amd-ryzen-7-1800x-processor-review,12.html

Ryzen should be doing better, pretty sure we got an issue Gamer and so forth think it could be SMT which sounds very feasible to me the benchmarks above would never allow Ryzen to test that high if IPC was real low.

I heard Linux/Unix should be getting a patch soon to.

I'll personally buy Ryzen once i get more $$$ and if this gets fixed. Something has to be wrong no other way around it, 6900K beats a 7700K In a decent amount of titles even when the 7700K is at 5Ghz Ryzen should be at least close to that as Ryan Shrout from PC perspective notes in his review

 

[-Fran-]

Or maybe the design with a partitioned LLC linked via a slow interconnect is just ridiculous to start and no magic scheduler will hide the deficiencies of that. I have been thinking on some other similar design and don't remember anything. Everyone, Intel, ARM, IBM, Sun,... uses unified LLC.

For practical purposes Ryzen is working as two quad-core CPUs glued together in a die.

Q6600 anyone?

In any case, Juan, it doesn't matter how awful or elegant a solution is; if it works, deal with it. Now, Intel has the better and more elegant solution it seems, but if it weren't the better performing, people wouldn't care.

The Phenom 1 (good ol' Barcie) was a "true" core and a very nice solution, but the Q6600 was better (and other Kentsfield successors), so no one really cared it was 2 dual cores slapped together.

I think you're missing the point with the CCX arrangement and it's a very clever one. I won't share my opinion, because it is plenty obvious and you might already even know why.

My point is not about "elegance", but about the performance deficits. Note as well that I was the first guy that predicted the quad-core clusters... I did this so early as 2014 when the codename Zen was not public still. I predicted that correctly because I was able to infer AMD modular plans as a way to reduce R&D costs. What I couldn't imagine is that they would do a quad-core modular approach with a split LLC and then get trapped into the difficulties they are experiencing now.

And "elegance" is also not my point; if it were better than Intel with no doubts, no one would say its a "bad" thing, but a "smart" thing, maybe?

AMD just chose that strategy (clusters) not for "performance" reasons, but business reasons. Like I said, AMD has to live in 3 market segments with a single uArch: server, consumer and multipart (think consoles). They had to define a "unit" that can be moved across designs easily and attack the important markets they can cash-in from. This is just a similar approach to the modules in BD, but a bit more clever and forward thinking (my take at least).

So, this is just a counterpoint to say it's not because the team "made a mistake" when choosing the CCX arrangement, but it was a conscious decision (from what I can see). I did like the "modules" approach in BD, but it didn't pan out how they wanted; now the CCX seems to be following the same principles that backed up the "module" approach in BD, but now done in a better way.

Cheers!

 

[truegenius]

juanrga said:
Or maybe the design with a partitioned LLC linked via a slow interconnect is just ridiculous to start and no magic scheduler will hide the deficiencies of that. I have been thinking on some other similar design and don't remember anything. Everyone, Intel, ARM, IBM, Sun,... uses unified LLC.

For practical purposes Ryzen is working as two quad-core CPUs glued together in a die.

why they used ccx instead of going octa ?
they are probabily using same die for 4 core part ( why ? ) too means yields have to be ridiculous ( or purposedly disable cores ) means even more performance loss and different perfromance for different core combination ( 4+0, 1+3, 2+2 )

btw is there any review in which they have tested by disabling cores to replicate 4/6c parts and predict their performance and to rule out ccx causing perfomance issue by comparing 4+0, 3+1, 2+2 config ?

 

[juanrga]

1 Ryzen die -> ~194mm^2
1 CCX -> 44mm^2

The die size is larger than that. Chip-architect estimation is incorrect. He got that number from measuring a die shot has been photoshoped by AMD.

The correct size is behind a paywall at SA, but I can say that the die is bigger than my old estimation of ~205mm^2...

Edit: The exact value is behind a paywall, but next link gives a very good hint to which is the real die size

https://twitter.com/TMFChipFool/status/839279240948367365?p=p

 

[juanrga]

Ryzen is 1% ahead of BDW-E in single thread, according to a conversation with Robert Hallock. They are 6.8% behind kaby lake, by their own admission.

Unless you think a 200-300 MHz advantage over Ryzen is going to magically produce 20% somehow when BDW-E is starting 1% behind at same clocks. I doubt a 4.1 GHz Ryzen loses by much at all to a 4.4 GHz 6900K in anything that is optimized.

The *true* performance increase will come from the April windows patch when the scheduler recognizes SMT cores. Microcode will clean up quite a bit of things, perhaps they pick up 5-10% in some areas...but the windows scheduler is what is really killing them at the moment.

AMD made several outstanding claims about performance (CB, Blender, gaming,...) that have been disproved by reviews. For instance both PCWorld and PCPer admit that cannot reproduce the CB scores. PCWorld: "My own tests don’t quite match AMD’s results." And PCPer writes: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated." GamerNexus even goes beyond and offer us an analysis of how AMD inflated gaming numbers:

At this point, you might be left feeling disillusioned when considering AMD’s tech demos. Keep in mind that most of the charts leaked and created by AMD revolved around Cinebench, which is not a gaming workload. When there were gaming workloads, AMD inflated their numbers by doing a few things:

In the Sniper Elite demo, AMD frequently looked at the skybox when reloading, and often kept more of the skybox in the frustum than on the side-by-side Intel processor. A skybox has no geometry, which is what loads a CPU with draw calls, and so it’ll inflate the framerate by nature of testing with chaotically conducted methodology. As for the Battlefield 1 benchmarks, AMD also conducted using chaotic methods wherein the AMD CPU would zoom / look at different intervals than the Intel CPU, making it effectively impossible to compare the two head-to-head.

And, most importantly, all of these demos were run at 4K resolution. That creates a GPU bottleneck, meaning we are no longer observing true CPU performance. The analog would be to benchmark all GPUs at 720p, then declare they are equal (by way of tester-created CPU bottlenecks). There’s an argument to be made that low-end performance doesn’t matter if you’re stuck on the GPU, but that’s a bad argument: You don’t buy a worse-performing product for more money, especially when GPU upgrades will eventually out those limitations as bottlenecks external to the CPU vanish.

Ryzen is a good 10% clock-for-clock behind BDW-E in applications. "moyenne applicative" means average of applications

And a good 20% behind BDW-E in games. "moyenne jeux 3D" means average of games

A BDW-E overclocked to 4.2--4.3GHz will increase the gap compared to a Ryzen overclocked to 3.9--4.0GHz.

New schedulers will not solve magically the muarch deficits. Just as they didn't solve for Bulldozer. I remember the multiple claims that Bulldozer scheduler was going to improve performance significantly. Then AMD and Microsoft did released two patches and average performance gains were 2--5%.

Note that above graphs already include most of the effect of a smart scheduler in the performance bars label "Core Parking OFF". I am using this better case for obtaining the above 10% and 20% gaps.

 

[juanrga]

juanrga said:
Or maybe the design with a partitioned LLC linked via a slow interconnect is just ridiculous to start and no magic scheduler will hide the deficiencies of that. I have been thinking on some other similar design and don't remember anything. Everyone, Intel, ARM, IBM, Sun,... uses unified LLC.

For practical purposes Ryzen is working as two quad-core CPUs glued together in a die.

why they used ccx instead of going octa ?
they are probabily using same die for 4 core part ( why ? ) too means yields have to be ridiculous ( or purposedly disable cores ) means even more performance loss and different perfromance for different core combination ( 4+0, 1+3, 2+2 )

btw is there any review in which they have tested by disabling cores to replicate 4/6c parts and predict their performance and to rule out ccx causing perfomance issue by comparing 4+0, 3+1, 2+2 config ?

I am preparing an article about my predictions about RyZen where I explain why I predicted what I did, where I was right and where I failed and why. It is still a draft, because I am waiting for Anandtech review 2. Regarding CCX I write:

I also predicted Zen would come in four-core clusters. This is what AMD names the CCX. My proposal was based in a hypothesis about AMD wanting to increase the minimal core count to four and a hypothesis about reusing the cluster for the APUs and for the semicustom division to reduce the design costs with a modular approach. My prediction was that Zen would come in groups of CCX with SMT disabled or not. For instance 4c/4t for the lowest Ryzen CPU and then 4c/8t and 8c/8t for the intermediate models, and 8c/16t for the flagship AM4 socket model. With servers coming in combinations of 8c/16t and 16c/32t for the dual die socket (SP4), and 16c/32t and 32c/64t for the quad die socket (SP3).

MOD EDIT: <LINK DELETED>

Reusing a basic compute unit of four-core helps AMD to reduce the R&D cost of design/testing/validating different configuration from mobile to server. You only have to design/validate/test the CCX once, and then reuse the CCX on different dies. For instance if tomorrow someone approach AMD for a semicustom desing. The customer only would chose the number of CCX it want. And AMd would only place the CCXs on a die and validate the interconnections between CCX and the rest of the SoC sving both money and time compared to a non-modular design from scratch.

My vision was quad-core modules but with a separate LLC module. The die, in my vision would be

[ compute-module ] <--> [ cache module (L3) ]<--> [ compute module ]

^^^^ I will prepare a proper diagram when have time. Meanwhile I expect the above is enough to clarify my vision.

With this design the L3 cache would be a true LLC for all cores. Surprisingly AMD chose a weird CCX approach with local L3 and connected to the other CCX (and its local L3) through a slow interconnect: only 22MB/s !!!!!!

I see in other forums people pretending that this weird configuration with split LLC is a consequence of RyZen being designed for servers, not games. That is an excuse. Power8 (IBM) or Vulcan (Broadcom) are designs exclusive for servers and both have unified LLC. Haswell/Broadwell/Skylake... are designs optimized for server and have unified LLC.

CPCHardware did some testing with engineering samples and found that 4+0, 3+1, and 2+2 have different performance. It seems commercial chips are constrained to having symmetric CCX or full disabled CCX. This means the next configurations:

6-core = 3+3
4-core = 4+0 or 2+2

 

[daerohn]

another issue is Intel will be releasing 8th gen core chips this year, which is said to be %15 faster than previous chips. while AMD has just released a new chip barely on par with intel 6th-7th gen, with the release of the new series AMD will be falling short again.

So AMD has again some rough times ahead. They will be strugling with devolopers, also will be releasing new drivers and BIOS updates and when they finally have a stable product ready for the market, will face a new opponent. This can cause AMD to loose its edge they have today.

Right now they may be selling a lot of product yet a lot more people like myself will be waiting to see the final results to make a purchase. I have been purchasing AMD chips since Athlon and I will again bu a ryzen chip in the next months but not before I see the platform is stabilized by means of operating system, BIOS and software updates.

If by this year you mean, in the next 15 months. then sure.

As it stands, Intel announced 8th Gen coming in Q2 2018.

the page below supposed it will be released 2nd half this year. Also there is a picture showing intel road map for 2017.

http://www.pcworld.com/article/3168827/components-processors/intels-8th-gen-coffee-lake-chips-reuse-14nm-process-as-other-core-cpus-ease-into-new-tech.html

infact I do not care for what Intel plans for this year. I haven't been purchasing Intel since 2004. and do not think to switch in near future.

 

[8350rocks]

Ryzen is 1% ahead of BDW-E in single thread, according to a conversation with Robert Hallock. They are 6.8% behind kaby lake, by their own admission.

Unless you think a 200-300 MHz advantage over Ryzen is going to magically produce 20% somehow when BDW-E is starting 1% behind at same clocks. I doubt a 4.1 GHz Ryzen loses by much at all to a 4.4 GHz 6900K in anything that is optimized.

The *true* performance increase will come from the April windows patch when the scheduler recognizes SMT cores. Microcode will clean up quite a bit of things, perhaps they pick up 5-10% in some areas...but the windows scheduler is what is really killing them at the moment.

AMD made several outstanding claims about performance (CB, Blender, gaming,...) that have been disproved by reviews. For instance both PCWorld and PCPer admit that cannot reproduce the CB scores. PCWorld: "My own tests don’t quite match AMD’s results." And PCPer writes: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated." GamerNexus even goes beyond and offer us an analysis of how AMD inflated gaming numbers:

At this point, you might be left feeling disillusioned when considering AMD’s tech demos. Keep in mind that most of the charts leaked and created by AMD revolved around Cinebench, which is not a gaming workload. When there were gaming workloads, AMD inflated their numbers by doing a few things:

In the Sniper Elite demo, AMD frequently looked at the skybox when reloading, and often kept more of the skybox in the frustum than on the side-by-side Intel processor. A skybox has no geometry, which is what loads a CPU with draw calls, and so it’ll inflate the framerate by nature of testing with chaotically conducted methodology. As for the Battlefield 1 benchmarks, AMD also conducted using chaotic methods wherein the AMD CPU would zoom / look at different intervals than the Intel CPU, making it effectively impossible to compare the two head-to-head.

And, most importantly, all of these demos were run at 4K resolution. That creates a GPU bottleneck, meaning we are no longer observing true CPU performance. The analog would be to benchmark all GPUs at 720p, then declare they are equal (by way of tester-created CPU bottlenecks). There’s an argument to be made that low-end performance doesn’t matter if you’re stuck on the GPU, but that’s a bad argument: You don’t buy a worse-performing product for more money, especially when GPU upgrades will eventually out those limitations as bottlenecks external to the CPU vanish.

Ryzen is a good 10% clock-for-clock behind BDW-E in applications. "moyenne applicative" means average of applications

And a good 20% behind BDW-E in games. "moyenne jeux 3D" means average of games

A BDW-E overclocked to 4.2--4.3GHz will increase the gap compared to a Ryzen overclocked to 3.9--4.0GHz.

New schedulers will not solve magically the muarch deficits. Just as they didn't solve for Bulldozer. I remember the multiple claims that Bulldozer scheduler was going to improve performance significantly. Then AMD and Microsoft did released two patches and average performance gains were 2--5%.

Note that above graphs already include most of the effect of a smart scheduler in the performance bars label "Core Parking OFF". I am using this better case for obtaining the above 10% and 20% gaps.

CB scores have been duplicated by various reviewers routinely.

The only one that has been errant is the 6900K, which in some cases has more variance.

Stock 1800X routinely produces 160-163 single thread in cinebench single core depending upon cooling solution allowing turbo for the full duration or not. Stock 6900K has produced between 155-161 typically in those same benchmarks.

Not earth shattering, but the 1% is there.

EDIT: It would also be prudent to point out to those viewing your graphs that they are comparing a 6950X in those benchmarks, and the gap you are referring to is from 2 more cores...or more prudently a 20% advantage in resources.

 

[salgado18]

I don't know if you guys saw this:

https://www.techpowerup.com/231268/amds-ryzen-cache-analyzed-improvements-improveable-ccx-compromises

It's a nice dive-in to the cache performance of Ryzen, compared to Broadwell-E and Piledriver. Basically, the cache is between equal to slower, but the problem arises when the interconnection comes into play, because the L3 cache is split between two modules, and the connection is much slower than the caches. Already said, but it's nice to have numbers.

 

[8350rocks]

https://www.youtube.com/watch?v=40h4skxDkh4

Details behind windows scheduling problem with Ryzen.

 

[juanrga]

Ryzen is 1% ahead of BDW-E in single thread, according to a conversation with Robert Hallock. They are 6.8% behind kaby lake, by their own admission.

Unless you think a 200-300 MHz advantage over Ryzen is going to magically produce 20% somehow when BDW-E is starting 1% behind at same clocks. I doubt a 4.1 GHz Ryzen loses by much at all to a 4.4 GHz 6900K in anything that is optimized.

The *true* performance increase will come from the April windows patch when the scheduler recognizes SMT cores. Microcode will clean up quite a bit of things, perhaps they pick up 5-10% in some areas...but the windows scheduler is what is really killing them at the moment.

AMD made several outstanding claims about performance (CB, Blender, gaming,...) that have been disproved by reviews. For instance both PCWorld and PCPer admit that cannot reproduce the CB scores. PCWorld: "My own tests don’t quite match AMD’s results." And PCPer writes: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated." GamerNexus even goes beyond and offer us an analysis of how AMD inflated gaming numbers:

At this point, you might be left feeling disillusioned when considering AMD’s tech demos. Keep in mind that most of the charts leaked and created by AMD revolved around Cinebench, which is not a gaming workload. When there were gaming workloads, AMD inflated their numbers by doing a few things:

In the Sniper Elite demo, AMD frequently looked at the skybox when reloading, and often kept more of the skybox in the frustum than on the side-by-side Intel processor. A skybox has no geometry, which is what loads a CPU with draw calls, and so it’ll inflate the framerate by nature of testing with chaotically conducted methodology. As for the Battlefield 1 benchmarks, AMD also conducted using chaotic methods wherein the AMD CPU would zoom / look at different intervals than the Intel CPU, making it effectively impossible to compare the two head-to-head.

And, most importantly, all of these demos were run at 4K resolution. That creates a GPU bottleneck, meaning we are no longer observing true CPU performance. The analog would be to benchmark all GPUs at 720p, then declare they are equal (by way of tester-created CPU bottlenecks). There’s an argument to be made that low-end performance doesn’t matter if you’re stuck on the GPU, but that’s a bad argument: You don’t buy a worse-performing product for more money, especially when GPU upgrades will eventually out those limitations as bottlenecks external to the CPU vanish.

Ryzen is a good 10% clock-for-clock behind BDW-E in applications. "moyenne applicative" means average of applications

And a good 20% behind BDW-E in games. "moyenne jeux 3D" means average of games

A BDW-E overclocked to 4.2--4.3GHz will increase the gap compared to a Ryzen overclocked to 3.9--4.0GHz.

New schedulers will not solve magically the muarch deficits. Just as they didn't solve for Bulldozer. I remember the multiple claims that Bulldozer scheduler was going to improve performance significantly. Then AMD and Microsoft did released two patches and average performance gains were 2--5%.

Note that above graphs already include most of the effect of a smart scheduler in the performance bars label "Core Parking OFF". I am using this better case for obtaining the above 10% and 20% gaps.

CB scores have been duplicated by various reviewers routinely.

The only one that has been errant is the 6900K, which in some cases has more variance.

Stock 1800X routinely produces 160-163 single thread in cinebench single core depending upon cooling solution allowing turbo for the full duration or not. Stock 6900K has produced between 155-161 typically in those same benchmarks.

Not earth shattering, but the 1% is there.

EDIT: It would also be prudent to point out to those viewing your graphs that they are comparing a 6950X in those benchmarks, and the gap you are referring to is from 2 more cores...or more prudently a 20% advantage in resources.

Precisely several reviewers have noted that cannot reproduce the CB numbers given by AMD and thatr the IPC is less than AMD promised

PCWorld: "My own tests don’t quite match AMD’s results."

PCPER: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated."

Also the above graphs are NOT for a 6950X.

Nous cadençons pour ce test nos FX-8350 (Piledriver), Ryzen 7 1800X (Zen) et Core i7 6900K (Broadwell-E) à 3 GHz.

It is 8-core Broadwell vs 8-core Ryzen, vs 8-core Piledriver. Ryzen is 10--20% slower than Broadwell because the IPC is well below Haswell and close to Sandy Bridge level.

 

[jaymc]

https://www.youtube.com/watch?v=40h4skxDkh4

Details behind windows scheduling problem with Ryzen.

Great video ... and also reassuring.. think the April update will do plenty for Ryzen.

I was thinking to myself, with all the moaning going on... I don't know about you guys but I'm still sold on it anyway...

I think most people are !

But still having a little moan about few frames per second here an there... I'm looking forward to seeing how AMD's sales of Ryzen are going, because I can't help thinking that even though the gaming community is complaining...
They are probly all doing it while ordering Ryzen anyway... regardless of the teething problems that is

 

[8350rocks]

Ryzen is 1% ahead of BDW-E in single thread, according to a conversation with Robert Hallock. They are 6.8% behind kaby lake, by their own admission.

Unless you think a 200-300 MHz advantage over Ryzen is going to magically produce 20% somehow when BDW-E is starting 1% behind at same clocks. I doubt a 4.1 GHz Ryzen loses by much at all to a 4.4 GHz 6900K in anything that is optimized.

The *true* performance increase will come from the April windows patch when the scheduler recognizes SMT cores. Microcode will clean up quite a bit of things, perhaps they pick up 5-10% in some areas...but the windows scheduler is what is really killing them at the moment.

AMD made several outstanding claims about performance (CB, Blender, gaming,...) that have been disproved by reviews. For instance both PCWorld and PCPer admit that cannot reproduce the CB scores. PCWorld: "My own tests don’t quite match AMD’s results." And PCPer writes: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated." GamerNexus even goes beyond and offer us an analysis of how AMD inflated gaming numbers:

At this point, you might be left feeling disillusioned when considering AMD’s tech demos. Keep in mind that most of the charts leaked and created by AMD revolved around Cinebench, which is not a gaming workload. When there were gaming workloads, AMD inflated their numbers by doing a few things:

In the Sniper Elite demo, AMD frequently looked at the skybox when reloading, and often kept more of the skybox in the frustum than on the side-by-side Intel processor. A skybox has no geometry, which is what loads a CPU with draw calls, and so it’ll inflate the framerate by nature of testing with chaotically conducted methodology. As for the Battlefield 1 benchmarks, AMD also conducted using chaotic methods wherein the AMD CPU would zoom / look at different intervals than the Intel CPU, making it effectively impossible to compare the two head-to-head.

And, most importantly, all of these demos were run at 4K resolution. That creates a GPU bottleneck, meaning we are no longer observing true CPU performance. The analog would be to benchmark all GPUs at 720p, then declare they are equal (by way of tester-created CPU bottlenecks). There’s an argument to be made that low-end performance doesn’t matter if you’re stuck on the GPU, but that’s a bad argument: You don’t buy a worse-performing product for more money, especially when GPU upgrades will eventually out those limitations as bottlenecks external to the CPU vanish.

Ryzen is a good 10% clock-for-clock behind BDW-E in applications. "moyenne applicative" means average of applications

And a good 20% behind BDW-E in games. "moyenne jeux 3D" means average of games

A BDW-E overclocked to 4.2--4.3GHz will increase the gap compared to a Ryzen overclocked to 3.9--4.0GHz.

New schedulers will not solve magically the muarch deficits. Just as they didn't solve for Bulldozer. I remember the multiple claims that Bulldozer scheduler was going to improve performance significantly. Then AMD and Microsoft did released two patches and average performance gains were 2--5%.

Note that above graphs already include most of the effect of a smart scheduler in the performance bars label "Core Parking OFF". I am using this better case for obtaining the above 10% and 20% gaps.

CB scores have been duplicated by various reviewers routinely.

The only one that has been errant is the 6900K, which in some cases has more variance.

Stock 1800X routinely produces 160-163 single thread in cinebench single core depending upon cooling solution allowing turbo for the full duration or not. Stock 6900K has produced between 155-161 typically in those same benchmarks.

Not earth shattering, but the 1% is there.

EDIT: It would also be prudent to point out to those viewing your graphs that they are comparing a 6950X in those benchmarks, and the gap you are referring to is from 2 more cores...or more prudently a 20% advantage in resources.

Precisely several reviewers have noted that cannot reproduce the CB numbers given by AMD and thatr the IPC is less than AMD promised

PCWorld: "My own tests don’t quite match AMD’s results."

PCPER: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated."

Also the above graphs are NOT for a 6950X.

Nous cadençons pour ce test nos FX-8350 (Piledriver), Ryzen 7 1800X (Zen) et Core i7 6900K (Broadwell-E) à 3 GHz.

It is 8-core Broadwell vs 8-core Ryzen, vs 8-core Piledriver. Ryzen is 10--20% slower than Broadwell because the IPC is well below Haswell and close to Sandy Bridge level.

You are quoting games not optimized for IPC measurements? I think the argument is flawed.

All the actual IPC benchmarks point to parity with broadwell. Using gaming benchmarks with flawed windows schedulers, microcode bugs and everything else currently at play is a flawed attempt to justify a bad prediction.

Once the scheduler issues and other things are fixed, we can discuss games. As it stands right now, Ryzen shows parity with broadwell *in spite of those flaws* in IPC driven benchmarks like cinebench. What do you think the benchmarks will look like when those things are corrected??

Ryzen is well past Sandy Bridge in IPC. There is not a valid argument to present that shows otherwise.

 

[8350rocks]

https://www.youtube.com/watch?v=40h4skxDkh4

Details behind windows scheduling problem with Ryzen.

Great video ... and also reassuring.. think the April update will do plenty for Ryzen.

I was thinking to myself, with all the moaning going on... I don't know about you guys but I'm still sold on it anyway...

I think most people are !

But still having a little moan about few frames per second here an there... I'm looking forward to seeing how AMD's sales of Ryzen are going, because I can't help thinking that even though the gaming community is complaining...
They are probly all doing it while ordering Ryzen anyway... regardless of the teething problems that is

I am building a 4K prosumer gaming rig for productivity and gaming.

Already have everything but the GPU picked out: https://pcpartpicker.com/list/cVX3BP

 

[James Mason]

But I've finally gotten everything ordered for my Ryzen build, and PCP FINALLY has all the parts I ordered actually on their site:

PCPartPicker part list / Price breakdown by merchant

CPU: AMD RYZEN 7 1700 3.0GHz 8-Core Processor (Purchased For $329.00)
CPU Cooler: MSI CORE FROZR L 71.3 CFM CPU Cooler (Purchased For $49.99)
Thermal Compound: Thermal Grizzly Kryonaut 1g 1g Thermal Paste (Purchased For $8.95)
Motherboard: MSI X370 GAMING PRO CARBON ATX AM4 Motherboard (Purchased For $179.99)
Memory: G.Skill TridentZ RGB 16GB (2 x 8GB) DDR4-3200 Memory (Purchased For $171.59)
Storage: Crucial MX300 275GB M.2-2280 Solid State Drive (Purchased For $89.99)
Storage: Western Digital BLACK SERIES 1TB 3.5" 7200RPM Internal Hard Drive (Purchased For $74.99)
Video Card: EVGA GeForce GTX 1080 8GB FTW Gaming ACX 3.0 Video Card (Purchased For $594.89)
Case: NZXT S340 (Black/Red) ATX Mid Tower Case (Purchased For $59.99)
Power Supply: SeaSonic G 550W 80+ Gold Certified Semi-Modular ATX Power Supply (Purchased For $69.99)
Case Fan: Thermaltake Riing 12 RGB Triple Pack 40.6 CFM 120mm Fans (Purchased For $56.99)
Total: $1686.36
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2017-03-08 18:52 EST-0500

 

[jaymc]

But I've finally gotten everything ordered for my Ryzen build, and PCP FINALLY has all the parts I ordered actually on their site:

PCPartPicker part list / Price breakdown by merchant

CPU: AMD RYZEN 7 1700 3.0GHz 8-Core Processor (Purchased For $329.00)
CPU Cooler: MSI CORE FROZR L 71.3 CFM CPU Cooler (Purchased For $49.99)
Thermal Compound: Thermal Grizzly Kryonaut 1g 1g Thermal Paste (Purchased For $8.95)
Motherboard: MSI X370 GAMING PRO CARBON ATX AM4 Motherboard (Purchased For $179.99)
Memory: G.Skill TridentZ RGB 16GB (2 x 8GB) DDR4-3200 Memory (Purchased For $171.59)
Storage: Crucial MX300 275GB M.2-2280 Solid State Drive (Purchased For $89.99)
Storage: Western Digital BLACK SERIES 1TB 3.5" 7200RPM Internal Hard Drive (Purchased For $74.99)
Video Card: EVGA GeForce GTX 1080 8GB FTW Gaming ACX 3.0 Video Card (Purchased For $594.89)
Case: NZXT S340 (Black/Red) ATX Mid Tower Case (Purchased For $59.99)
Power Supply: SeaSonic G 550W 80+ Gold Certified Semi-Modular ATX Power Supply (Purchased For $69.99)
Case Fan: Thermaltake Riing 12 RGB Triple Pack 40.6 CFM 120mm Fans (Purchased For $56.99)
Total: $1686.36
Prices include shipping, taxes, and discounts when available
Generated by PCPartPicker 2017-03-08 18:52 EST-0500

@8350rocks Interested to know what you are getting out of your Liquid Cooler... quite reasonable at $125 nice.

@james mason Also nice build mate... notice you went with the 1700...Overclocked to 1800x speed I presume.

Can't wait to build one I'm going to donate my current 8350 to my children with an rx480 in it. An I'm confident they will still get plenty of enjoyment out of the old dog.!!

MOD EDIT: Off-topic references removed from this and prior posts

 

[TMTOWTSAC]

I'm looking forward to seeing how Ryzen performs with the 1080 ti, and especially Vega. On the one hand, as GPU's get faster it should create a performance gap at high resolutions. But on the other, most recent comparisons have showed lower CPU usage with AMD cards against comparable Nvidia cards. But there hasn't really been anything to compare the 1080 and Titan to on the AMD side. If that trend holds true, it could provide a significant boost to Ryzen in high end gaming, assuming Vega meets expectations.

 

[Nope 1151]

Juan, would AMD have been better off by splitting the L3 cache and using the actual ram sticks (L4?) as the "unified cache"? Could you explain the pro/cons (Aside from the obvious, "Less cache").

The way I see this is that AMD chose to aim for the server market, with good reason. The last thing Intel wants is for a company to say that they can easily replace Intel systems with AMD ones.

Competition is good.
Also, Mini-itx yay!
See this:
https://smallformfactor.net/news/biostar_racing_x370-gtn_itx

 

[dgothi]

Spoiler

Spoiler

AMD asked them to publish 1080, 1440, and 4K reviews together. They did not ask them to *not* do 1080p, they just asked them to show higher resolutions. This was clarified by Lisa Su in the reddit AMA and was characterized as Steve making a gross uncharacterized jab.

Other reviews confirm GN claims. For instance pcgamer claims: "AMD's first suggestion was to test at 1440p or 4K, which is complete bunk. Testing higher resolutions will absolutely put Ryzen's gaming performance closer to on par with Intel, but only in the sense that running higher resolutions shifts the bottleneck to the GPU. Even AMD's FX-series performs relatively close to Intel in many games, provided you're running at 4K. But if you want to know how Ryzen compares to Core i5/i7 when the GPU isn't the bottleneck, you need to test at lower resolutions."

Note that AMD didn't suggest them to test both 1080p and 4K (or 1440p), but to test only at 4K (or 1440p) to hide the problem by generating a GPU bottleneck.

Also, as noted by GN, AMD public demos of gaming on Ryzen were run exclusively at 4K generating a GPU bottleneck and hidding the problem to everyone.

Additionally..Steve from GN published a review stating that GPUs are better for video encoding. Except that they are not used for professional video encoding...at all.

GN is rigth. Funny enouh it is the own AMD which has been saying us for years that GPUs are better for that kind of workloads. I still recall when AMD announced to everyone that tasks as Handbrake would run faster on a GPU than in a CPU. The reason is that GPUs are more optimized for throughput whereas CPUs are more optimized for latency. This is also the basis for HSA (Heterogenenous System Architecture) which identifies CPUs with LCUs (Latency Compute Units) and GPUs with TCUs (Throughput Compute Units).

Steve also contradicted himself in his review...he ignored the minimum FPS where Ryzen was running neck and neck with all the Intel parts, but the maximum FPS was lower.

I saw him discussing minimum FPS also and they were lower than in an i5 in several titles.

Additionally, they tested in DX12 exclusively, used odd blender settings, and other weird methodology. Joker tested in DX11, and his results are reflected by other DX11 tests.

Not true. GN also tested DX11 titles. Joker tests were unprofessional (large marging of error) and irrelevant because he chose odd settings that generated a GPU bottleneck with the GPU at 99% of load.

This reddit nitpick of GN's review pretty much sums it up: https://www.reddit.com/r/Amd/comments/5xgonu/analysis_of_gamersnexus_r7_1800x_review_an_i5_in/

GN conclusions agree with all other reviews that also found that Ryzen is not good at gaming. There are a fair good consensus among reviewers.

PcGamer: "The AMD Ryzen 7: plenty of power, but underwhelming gaming".

Arsctechnica: "an excellent workstation CPU, but it doesn't game as hard as we hoped".

Extremetech: "Zen is an amazing workstation chip with a 1080p gaming Achilles heel"

And of course TomsHardware: "It’s hard to recommend the Ryzen 7 1800X over Intel's lower-cost quad-core chips for gaming, especially given the Core i7-7700K's impressive performance."

I think the best comment on that reddit thread was made by the user that writes: "aka: Joker got the biased results we wanted out of hundreds of other reviewers and that is why we like him and are throwing GN and others under the bus".

All of this leads me to a few conclusions regarding the launch:

The BIOS was very buggy, and this was reflected in numerous Asus reviews, Gigabyte seemed to fare much better overall in that regard, especially in regards to memory timings and higher memory clock support.

Problems have been reproduced on non-Asus mobos.

The CCX cross threading is a major issue, supposedly windows is patching it next month, we will see. Windows 7 posts gaming numbers for Ryzen that are anywhere from 10-16% better according to the Stilt. Performance in other areas sees a significant improvement as well.

The CCX approach with a split LLC is weird. It is AMD fault, not everyone else. In any case, I don't expect large gains from patching Windows. Recall the Bulldozer scheduler, it improved performance by about 2--5%.

Agree 100% nicely said. I actually expected Fans to react this way but i never ever expected Amd to do this stuff i really didn't i thought that type of stuff was in the past(Bulldozer days).

Eitherway if you can watch Jayz latest video he did show that Ryzen is not being reported right in heaven benchmark and heaven benchmark was only using 1 core. If games are doing something like that it could be a reason why. I mean Ryzen does really well in everything but gaming that is why i find the results to be odd.

" No one talks about Cinebench scores anymore since Ryzen is now beating its counterpart there" I do Ryzen is great for everything but gaming where its mediocre.

"IS NOT RYZEN'S GAMING LINE OF PROCESSORS"

Then what will be the 4 and 6 core parts will look worse and i doubt they will be clocked higher then 300Mhz over the 1800X. 6 core part will be aimed at I5 prices where the I5 can be OC to 4.5Ghz easily, the 4 core part will look better being aimed at the I3 and i do think budget gamers will love that part.

From various comments you just made, are you telling me that if someone came to you and said all I do is video game, its all I'm interested in and I want to build the best video gaming system possible then you would hose them and building a system around the i7 6900K??? You wouldn't be in business for very long doing that to customers and then quote a couple outlier games where the i7 6900K beast the i7 7700K. Lets be honest, if someone asked you for gaming system you would recommend the i7 7700K over the i7 6900K for gaming as most of the time it has better gaming performance - and that isn't even taking into account the insane price Intel puts on its i7 6900K. Right now 8 core 16 thread processors are not the gaming line of processors, they are the workstation class of processors.

The gaming processors are the 4 core 8 thread and maybe even 6 core 12 thread processors. The gaming processors for Ryzen will be the R5 4 and 6 core processors, and they should boast better overclocking potential than their massive R7 brothers. As many others have stated as well Ryzen's gaming issue(s) aren't Bulldozer's issue. Bulldozer's low gaming performance was easy to understand from the get go, all one had to do was run any benchmark like Cinebench, look at the very low single core performance and bingo that is why it doesn't game in the enthusiast market. Ryzen is totally different, from all the benchmarks we have, with the powerhouse it is it should be gaming much, much better. Something is currently hobbling it, but it is far from the hardware issues Bulldozer had. Ryzen has very good IPC and very good single core execution, once the issue is found that is crippling its 1080p gaming potential it should hit the same gaming benchmarks as Skylake based on its very impressive workstation performance.

R5 Ryzen processors will bring AMD back into enthusiast gaming, but they won't dominate there, where they have a chance of dominating is with their R7 and Naples line in workstation and server application.

No i'd have them get a 7700K and according to IPC test's and frequency scaling on Ryzen OC i'd still recommend the 7700K as the best gaming processor 300mhz more on ryzen won't change that.

R5 will not bring Amd back in Enthusiast gaming it will make them a option again for people who do more then gaming. R3 will be Amd's major win CPU.

i5s are not even the recommended gaming processor anymore. Seriously...4 core i7s are what people are recommending for high end gaming.

As for enthusiast gaming, that depends, does it not? I intend to go 4K this time around, and I may run CF Vega GPUs to do 3 monitors...(still undecided about multi-monitor...to be honest...)

1440p Ryzen is fine, the differences are not substantial enough to warrant paying more for a 7700k. If you only play at 1080p...then maybe you look at a 7700k. I would only say sure if literally all you did was gaming, not capture software, no streaming, and no additional apps in the background. Simply because the 7700k is very nearly bottlenecking at 1080p in many games, and Ryzen has enough cores it would not impact performance in the same way.

Additionally...the 1% and 0.1% low FPS numbers are typically in favor of Ryzen, meaning a much smoother overall experience in spite of a 15-20% deficit in frames compared to a 5 GHz 7700k.

You said,"i5s are not even the recommended gaming processor anymore."

Intel is going to release new Intel 8th generation so i5 will have hyper-threading feature later this year. that is what I read it on Intel's plan on news. I assume i5 will be back with recommendation for gaming. But i5-7600k still strong and long time for 1080/1440p with decent graphic card. Also, I believe Ryzen 5 will challenge i5.

-DG

No, i5s will not receive hyper threading. They will just up core counts to 6 cores supposedly on 8th Gen in 2018 when it launches in Q2.

i5 get 6 cores is a possible.it will be cost more because it is similar to Intel's 5820k 6 cores however, 5820k have hyperthreading. I assume i5 might get 6 cores OR 4 cores with hyperthreading either. we will wait and see 🙂

 

[dgothi]

people who bought Ryzen 1700 so it received very well review on amazon, I am surprise. I have not decide which do I want to buy intel or Ryzen. No rush. Intel i7-7700 might be safe to buy? Maybe since it is maturity, right? I might buy either Ryzen or Intel next two weeks.
-DG

 

[TechyInAZ]

people who bought Ryzen 1700 so it received very well review on amazon, I am surprise. I have not decide which do I want to buy intel or Ryzen. No rush. Intel i7-7700 might be safe to buy? Maybe since it is maturity, right? I might buy either Ryzen or Intel next two weeks.
-DG

If you game a lot, the i7-7700 might be better, but if you do anything else, Ryzen is the way to go.

 

[juanrga]

You are quoting games not optimized for IPC measurements? I think the argument is flawed.

All the actual IPC benchmarks point to parity with broadwell. Using gaming benchmarks with flawed windows schedulers, microcode bugs and everything else currently at play is a flawed attempt to justify a bad prediction.

Once the scheduler issues and other things are fixed, we can discuss games. As it stands right now, Ryzen shows parity with broadwell *in spite of those flaws* in IPC driven benchmarks like cinebench. What do you think the benchmarks will look like when those things are corrected??

Ryzen is well past Sandy Bridge in IPC. There is not a valid argument to present that shows otherwise.

Game benches, if well made can measure IPC. "Well made" means one is not introducing GPU bottlenecks of any class that artificially reduce the CPU performance. Example of bad measurement:

The settings used have generate a GPU-bottleneck and all the CPUs perform the same because all them are waiting for the GPU to do its graphic work.

Good measurement showing CPUs real performance differences:

But let us ignore games and check only applications. FX-8350 (8-core Piledriver) vs Ryzen 7 1800X (8-core Zen) vs i7 6900K (8-core Broadwell-E) all them at 3 GHz

Even with a smart scheduler that reduces SMT penalty on Zen, we can check that Broadwell is faster than Zen. In some specific workload as x264 Zen is slightly ahead , but in others like WinRAR Zen is clearly behind. Overall ("moyenne applicative") Broadwell is a good 10% ahead of Zen on average.

A 10% behind Broadwell put Zen at the Sandy Bridge level more or less. This was expected because the Zen microarchitecture is inferior to Haswell. Reviews finally confirmed this.

Cinebench shows the same pattern than the above average over applications. The marketing slides given by AMD don't reflect reality, as reviewers have confirmed:

PCWorld: "My own tests don’t quite match AMD’s results."

PCPER: "Still, the 8% gap between the 6900K and the Ryzen 7 1800X at 3.5 GHz tells me that AMD’s claims of equal IPC appear to have been overstated."

This is PCPER measurement of IPC using CineBench

Ryzen performs like Ivy Bridge more or less. On Handbrake the IPC od Ryzen is better and matches Broadwell

but on Audacity the IPC of Ryzen is much worse than Broadwell

On workstation-class applications the IPC of Zen is about 10% lower than Broadwell. On games the IPC of Zen is about 20% lower than Broadwell. Overall, a 15% below Broadwell is Sandy Bridge territory.