AMD Ryzen 2 Review Mega Thread

[ZestyJalapeno]

AMD you absolute fucking unit. Time to sell the 4790K

[SpaceGhostC2C]

6 hours ago, ImNotDeViLzzz said:

WTB reviews with 4K gaming benchmarks.  Seeing them use a GTX 1080 Ti and these cpus and not doing 4K gaming benchmarks is ridiculous. So far all I have seen is reviews with 1080p and 1440p gaming.

They just want to make the GPU overkill so it's all CPU-bound (sometimes they still fail at 1080p).

But go over to Anandtech and you'll find 4K for anything, and in Civilization VI's case even 8K and 16K (which is frankly a waste of time). Because, yes, they tested Civ VI in terms of FPS, not turn time... (they also explain why, though).

[ImNotDeViLzzz]

14 minutes ago, SpaceGhostC2C said:

But go over to Anandtech and you'll find 4K for anything ...

Thanks so much.  They had exactly what I was looking for. : )

[Humbug]

4 hours ago, Tic-Tac said:

Stop focusing on if Anandtech destroyed Coffee Lake's performance. They didn't. Look back at their coffee lake review and all the game numbers are the same. The real question is, how did they get Ryzen to perform so well!

Hmm interesting. And in the coffee lake review the coffee lake parts beat 1st gen Ryzen parts in gaming as expected.

 

So now the conversation should shift to whether or not he can reliably reproduce such high gaming performance on 2nd gen ryzen. Or did he mix up some numbers.. 

[Brooksie359]

9 hours ago, DoctorWho1975 said:

 

If you are streaming a stock Ryzen 2 destroys the 8700k, even when you overclock the 8700k, see GN video for a comparison, its pretty startling.

To be fair that was mostly at higher quality. When at 10 mbs it was pretty even but at 12 it was not. So it really depends on what you plan to do for quality settings. 

[Brooksie359]

9 hours ago, Misanthrope said:

I don't disagree with any of that, in fact I think it pretty much covers the use case for a 2700(x) instead of an 8700: an amateur streamer/content creator or someone who otherwise has a passing interest on it.

 

But yeah I didn't really address it as "this isn't a niche" but more among the lines of "Within the people who might stream, a respectable percentage will want an HEDT instead" so it's not automatically all streamers are 2700 there are better options for many of them.

If you look at the HEDT 8 core compared to the 2700x in hardware unboxed video it shows that they are basically the same performance give or take but the 2700x is much cheaper. If you want to spend alot of money on a streaming setup you will want a 2 system setup. 

[Taf the Ghost]

3 hours ago, Humbug said:

Hmm interesting. And in the coffee lake review the coffee lake parts beat 1st gen Ryzen parts in gaming as expected.

 

So now the conversation should shift to whether or not he can reliably reproduce such high gaming performance on 2nd gen ryzen. Or did he mix up some numbers.. 

Anandtech's Intel scores actually look pretty reasonable, compared to their launch tests, it's the Ryzen 2x00 results that seem higher than others. Some of Anandtech's gaming results always come out a little odd (probably down to quality choices within the game), but I think they might have a version of the Sleep Bug in a few of their tests, which would explain the issue.

 

Though: https://www.techradar.com/reviews/amd-ryzen-7-2700x

 

We're starting to get tests with the Spectre v2 fixes, though this seems to be the most extreme. (Which probably means a bad BIOS on the board being used for the 8700k) Which brings up a fascinating sub-plot to this story: is the Intel fixes starting to really hurting high FPS gaming?

[asus killer]

basically Ryzen 2 is what we all expected, nothing more, nothing less (well i guess the coolers thing is news). People go nuts over nothing. Now it's GN review for streamers.

Need more threads go with Ryzen, anything else including gaming go Intel. Nothing to see here folks, move along.

 

Next totally unexpected move will be: Intel releasing a new processor that's just a little better than 8770k to crush the AMD's fans hopes and with a new chipset of course (not because they love to milk the idio... pardon customers, but because it must be done for reasons and stuff). This time not as rushed because we all saw AMD's move coming miles away (including Intel)

[Trixanity]

Is it possible that the Ryzen lead is simply caused by memory speeds? While I wasn't aware of their being such a big discrepancy, from what I can tell Intel runs at 2666 and Ryzen at 2933 (they run JEDEC speeds in all benchmarks if I recall). Seems far-fetched to me but that coupled with other factors (such as golden sample with a particularly good board and bios for better boost speeds) could give them a slight edge. 

 

If I recall TPU benchmarks said that CFL is faster than Ryzen by 7% at 1080p, 3% at 1440p and 1% at 2160p. However most other benchmarks use XMP speeds and maybe it was just the leaks but I saw some with Ryzen running 3200 and CFL running 3600. Combine that with questionable security patch levels and that may explain the discrepancy.

 

Well, it may also have been Anandtech mixing up numbers or fucking up the benchmarks themselves. I mean Ian Cutress did say he had been up for two days and had to scrap 36 hours of benchmarks due to a inadvertently disabled PB2.

[Taf the Ghost]

1 hour ago, asus killer said:

basically Ryzen 2 is what we all expected, nothing more, nothing less (well i guess the coolers thing is news). People go nuts over nothing. Now it's GN review for streamers.

Need more threads go with Ryzen, anything else including gaming go Intel. Nothing to see here folks, move along.

 

Next totally unexpected move will be: Intel releasing a new processor that's just a little better than 8770k to crush the AMD's fans hopes and with a new chipset of course (not because they love to milk the idio... pardon customers, but because it must be done for reasons and stuff). This time not as rushed because we all saw AMD's move coming miles away (including Intel)

I honestly think the Stock vs Stock increases are a little better than expected, especially given that the R5 1600 was a big seller. It's not enough to generally suggest an upgrade, unless you got some really bad silicon.

32 minutes ago, Trixanity said:

Is it possible that the Ryzen lead is simply caused by memory speeds? While I wasn't aware of their being such a big discrepancy, from what I can tell Intel runs at 2666 and Ryzen at 2933 (they run JEDEC speeds in all benchmarks if I recall). Seems far-fetched to me but that coupled with other factors (such as golden sample with a particularly good board and bios for better boost speeds) could give them a slight edge. 

 

If I recall TPU benchmarks said that CFL is faster than Ryzen by 7% at 1080p, 3% at 1440p and 1% at 2160p. However most other benchmarks use XMP speeds and maybe it was just the leaks but I saw some with Ryzen running 3200 and CFL running 3600. Combine that with questionable security patch levels and that may explain the discrepancy.

 

Well, it may also have been Anandtech mixing up numbers or fucking up the benchmarks themselves. I mean Ian Cutress did say he had been up for two days and had to scrap 36 hours of benchmarks due to a inadvertently disabled PB2.

Anandtech's results seem to be extra good from Ryzen. Techradar has the bad Coffee Lake results.

 

As for how the processors stack up, at "Stock" settings, AMD actually got a rather sizable improvement in memory latency, so the performance compares a lot better. It's actually at higher memory speeds where Intel can open up a nice gap with a 1080 Ti. (Anandtech's benchmarks are with a 1080.)

[Misanthrope]

4 hours ago, Brooksie359 said:

If you look at the HEDT 8 core compared to the 2700x in hardware unboxed video it shows that they are basically the same performance give or take but the 2700x is much cheaper. If you want to spend alot of money on a streaming setup you will want a 2 system setup. 

I was pretty clearly talking about the 10+ core chips not the entry level HEDTs because yes: I agree those are fucking stupid, both intel and AMD 8 core HEDTs.

[PrivateBits]

First... I'm a fan of AMD.  Always have been.

BUT, I feel the overwhelming need to point out that... well.. these new CPUs are almost laughable.  Why anyone would invest in Zen+ is simply beyond me, especially with the dramatically higher heat and power consumption for such tiny performance gains.  

If you are in the market for a new system, even then I'd probably NOT recommend these for almost anyone.  There just isn't a good price/performance/life expectancy target to hit with this final gen product (not Zen2).

I'm not trying to be a jerk; but I need to say at amplified volume.  Buying this product is .. well... just foolish and I'm really disappointed with Linus for not saying that himself.  Someone is a sell out.  You guess who for yourself.

Is there a market for this product.  Maybe.  Is this seriously a "next-gen" product?  Not really, not really at all.  Is this leading to advances that will be found in upcoming AMD solutions?  Again, not really.  Was this launch a complete waste of AMD's valuable time and energy.  Yeah, it really was and probably this launch did more to hurt AMD than anything else because it has no real impact on the growth of either the product line or the company as everyone knows that Zen2 is the next advancement target to hit.  This was more like just slapping some lipstick on the Ryzen and changing some packing, nothing more.   

Sorry... it is what it is.  I for one am not interested in a CPU that actually draws more power, runs 20 percent hotter and where the CPU will have a shorter life cycle and higher failure probability because of those issues.  Not really my bag.  

I'm not saying to run out and buy an Intel CPU (necessarily); but I am saying that if you are waiting for a real advancement, one where the capabilities and performance numbers are significantly improved, this isn't the solution you should be looking at.

Sorry, I'm not lying for anyone, even AMD.

[Humbug]

Oh dear

[leadeater]

10 minutes ago, PrivateBits said:

-snip-

More performance = more heat. Not unexpected. This was never touted as next generation anything either, not by AMD. All this was supposed to be was an improvement to the memory controller (check) and improvements to multi core boost (check) and better power efficiency per clock (also check).

 

The overall increase in total CPU power is from more cores being allowed to boost higher at once, this may look like worse power efficiency but that is not the case.

 

Also the extra heat will not have an impact of CPU life, not hot enough and CPUs also thermal throttle as well. I mean heck I've got a reference, yes reference, 6970 that's been used non stop since 2010 that sits at 90C during gaming and was in a system that never got powered off for 5 of those years. Now the GPU is in a friends computer, his 570 died, and shows no signs of degradation.

 

Faulting VRMs kills CPUs, not heat.

[ravenshrike]

16 minutes ago, PrivateBits said:

First... I'm a fan of AMD.  Always have been.

BUT, I feel the overwhelming need to point out that... well.. these new CPUs are almost laughable.  Why anyone would invest in Zen+ is simply beyond me, especially with the dramatically higher heat and power consumption for such tiny performance gains.  

If you are in the market for a new system, even then I'd probably NOT recommend these for almost anyone.  There just isn't a good price/performance/life expectancy target to hit with this final gen product (not Zen2).

I'm not trying to be a jerk; but I need to say at amplified volume.  Buying this product is .. well... just foolish and I'm really disappointed with Linus for not saying that himself.  Someone is a sell out.  You guess who for yourself.

Is there a market for this product.  Maybe.  Is this seriously a "next-gen" product?  Not really, not really at all.  Is this leading to advances that will be found in upcoming AMD solutions?  Again, not really.  Was this launch a complete waste of AMD's valuable time and energy.  Yeah, it really was and probably this launch did more to hurt AMD than anything else because it has no real impact on the growth of either the product line or the company as everyone knows that Zen2 is the next advancement target to hit.  This was more like just slapping some lipstick on the Ryzen and changing some packing, nothing more.   

Sorry... it is what it is.  I for one am not interested in a CPU that actually draws more power, runs 20 percent hotter and where the CPU will have a shorter life cycle and higher failure probability because of those issues.  Not really my bag.  

I'm not saying to run out and buy an Intel CPU (necessarily); but I am saying that if you are waiting for a real advancement, one where the capabilities and performance numbers are significantly improved, this isn't the solution you should be looking at.

Sorry, I'm not lying for anyone, even AMD.

Wut? First of all, it's running XFR 2 which means that before it would thermal throttle it backs off.in a controlled fashion to maximize performance. Same reason you see the increase in power consumption. I mean, if you were to forcibly clock the 2000 series at 4 Ghz, they would use as much power as the 1000 series at 3.7-3.8 Ghz. Instead it's hitting at least 4.2 pretty much all the time without having to manually dick around with things. Secondly, it allowed them to run a team whose only job was tightening down slack in memory latency and test the improvements in the real world without having to deal with a full node shrink and other major improvements to the architecture at the same time.

[cj09beira]

21 minutes ago, PrivateBits said:

 

sorry to say this but you are wrong, this is very important to amd, first because it gives them a much needed refresh of the bench data which means that optimizations and bios improvements made can be seen, secondly the better boost means that for the average joe that doesn't overclock this is quite a bit faster than if they had gotten ryzen 1, and then it helps keeping the ryzen name fresh in peoples heads.

sure its not a huge improvement but thats not a bad thing after all its been only a year so there isn't too much they could have done, and even then the improvements to the cache for example are quite impressive 

[Taf the Ghost]

32 minutes ago, leadeater said:

More performance = more heat. Not unexpected. This was never touted as next generation anything either, not by AMD. All this was supposed to be was an improvement to the memory controller (check) and improvements to multi core boost (check) and better power efficiency per clock (also check).

 

The overall increase in total CPU power is from more cores being allowed to boost higher at once, this may look like worse power efficiency but that is not the case.

 

Also the extra heat will not have an impact of CPU life, not hot enough and CPUs also thermal throttle as well. I mean heck I've got a reference, yes reference, 6970 that's been used non stop since 2010 that sits at 90C during gaming and was in a system that never got powered off for 5 of those years. Now the GPU is in a friends computer, his 570 died, and shows no signs of degradation.

 

Faulting VRMs kills CPUs, not heat.

 

24 minutes ago, ravenshrike said:

Wut? First of all, it's running XFR 2 which means that before it would thermal throttle it backs off.in a controlled fashion to maximize performance. Same reason you see the increase in power consumption. I mean, if you were to forcibly clock the 2000 series at 4 Ghz, they would use as much power as the 1000 series at 3.7-3.8 Ghz. Instead it's hitting at least 4.2 pretty much all the time without having to manually dick around with things. Secondly, it allowed them to run a team whose only job was tightening down slack in memory latency and test the improvements in the real world without having to deal with a full node shrink and other major improvements to the architecture at the same time.

Guys, that's clearly either a paid shill or a paid bot.

[leadeater]

3 minutes ago, Taf the Ghost said:

Guys, that's clearly either a paid shill or a paid bot.

Still worth checking to make sure. I don't mind wasting a few keyboard presses.

[cj09beira]

2 minutes ago, Taf the Ghost said:

 

Guys, that's clearly either a paid shill or a paid bot.

hope so

[Taf the Ghost]

1 minute ago, leadeater said:

Still worth checking to make sure. I don't mind wasting a few keyboard presses.

I've actually started to see instances of real people, not shills, voicing themselves in the similar manner of shills because it's become such a common comment style that normal people replicate, thinking it's the way you're supposed to discuss things. It's a sad state we've entered, but the ability for hyper fast web-crawlers and the ability to basically auto-generate non-repetitive talking points has really changed the nature of forums.

[asus killer]

3 hours ago, Taf the Ghost said:

I honestly think the Stock vs Stock increases are a little better than expected, especially given that the R5 1600 was a big seller. It's not enough to generally suggest an upgrade, unless you got some really bad silicon.

ad that to the expected Intel response and nothing changes. You'll be hearing AMD did great but Intel is the way to go and the 9700k or something will be king.

Real change will come with Zen 2 next year, that may finally level the playing field.

[Blademaster91]

25 minutes ago, Taf the Ghost said:

 

Guys, that's clearly either a paid shill or a paid bot.

Are they? I dunno the wording does seem odd, then again this is the first forum ive been on where spam bots are a occasional problem, requiring noobs to wait 5-10mins for a confirmation email before posting would maybe help but l'm getting OT.

There is kind of a point though, everyone praises AMD for the smallest gain while ryzen+ is rather like going from z170 to z270 an incremental upgrade and hyping it up to get you buying another board with really no new features. When Intel gives you a 2-7% incremental bump everyone screams ''OMG GREEDY SHINTEL WHERES THE REAL UPGRADE"?!?!!!ONE

[cj09beira]

12 minutes ago, Blademaster91 said:

Are they? I dunno the wording does seem odd, then again this is the first forum ive been on where spam bots are a occasional problem, requiring noobs to wait 5-10mins for a confirmation email before posting would maybe help but l'm getting OT.

There is kind of a point though, everyone praises AMD for the smallest gain while ryzen+ is rather like going from z170 to z270 an incremental upgrade and hyping it up to get you buying another board with really no new features. When Intel gives you a 2-7% incremental bump everyone screams ''OMG GREEDY SHINTEL WHERES THE REAL UPGRADE"?!?!!!ONE

stock for stock it really isnt though 

[Razor01]

 

 

There is a big shift in power efficiency in both single threaded and multi threaded work loads for Ryzen 2

[leadeater]

11 minutes ago, Razor01 said:

There is a big shift in power efficiency in both single threaded and multi threaded work loads for Ryzen 2

Gamers Nexus info on this is good too.

 

Quote

Blender V-F | R7 2700X vs. 1700
EFI Input VCore	1.175V (LLC lvl 4) 2700X	1.425 (LLC lvl 5) 1700
HWINFO VCore SV12 TFN	1.162v	1.425v
Motherboard VCore	1.145v	1.406v
Tdie at 15 minutes	57.8*C	79.4*C
Current Clamp Amperage	10.5A	16.3A
Clamp Calculated Wattage	129.15	200.49
CPU Core Current SV12 TFN	78A	101-108A
Ambient @ 15 mins	25.4*C	27.8*C

 

Quote

To this end, we found that, at a given frequency of 4.0GHz, our R7 2700X held stable at 1.175V input at LLC level 4, which equated to 1.162V VCore at SVI2 TFN. The result was stability in Blender and Prime95 with torturous FFTs, while measuring at about 129W power consumption in Blender. For this same test, our 1700 at 4.0GHz required a 1.425V input at LLC level 5, yielding a 1.425VCore, a 201W power draw – so 70W higher – and pushed thermals to 79 degrees Tdie. That’s up from 57.8 degrees Tdie at the same ambient.

 

https://www.gamersnexus.net/hwreviews/3287-amd-r7-2700-and-2700x-review-game-streaming-cpu-benchmarks-memory?showall=1

 

The high power 2700X is a fair decent amount less power draw compared to the previous low power 1700.