AMD Ryzen (Summit Ridge) Benchmarks Thread (use new thread)

[Doom2pro]

is there any reason for AMD to not reveal their launch lineup right now when retail listings are out?

It would be enforcing bad behavior on the part of the leakers, as I assume AMD didn't want this information out yet.

 

[CentroX]

It would be enforcing bad behavior on the part of the leakers, as I assume AMD didn't want this information out yet.

AMD is dreaming if they dont think SKU listings appear in advance. They always do.

 

[Doom2pro]

AMD is dreaming if they dont think SKU listings appear in advance. They always do.

They may also be subject to change, I don't know if any product has swapped hands yet... Who knows maybe these places already received stock.

 

[lolfail9001]

is there any reason for AMD to not reveal their launch lineup right now when retail listings are out?

They want to make it a hard launch.

 

[CatMerc]

They want to make it a hard launch.

It will not be the only hard thing at launch...




I'm talking about Intel's response of course

 

[CentroX]

It will not be the only hard thing at launch...




I'm talking about Intel's response of course

Intels response is in august

AMD has 4-5 months to gain marketshare. Not bad at all.

 

[sm625]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

 

[CentroX]

I dont get it. Why is that cheap cpu beating Skylake? Not concerned.

Also what was the clockspeed?

 

[Doom2pro]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

Yet the SKUs listed lately all start with YD not ZD... Last one I seen with ZD was a Bristol Ridge APU engineering sample: AMD Eng Sample: ZD3101AHM44AB_38/31/13/07_9874 with Radeon R7

Wouldn't be surprised if that was a Bristol Ridge part, or faked.

 

[zinfamous]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

your 3 year-old $50 CPU beating a 1 year-old $340 high-end Skylake? ...why would anyone ever buy 6700k? What a crap CPU!

lol crap--beaten by centrox

 

[Doom2pro]

Also there is a supposed Zen engineering sample picture going around with that same code, probably a Bristol Ridge AM4 part (thus the blacked out parts).

https://www.pic-upload.de/view-32649656/RyzenSampleRevision6.png.html

 

[majord]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

if those results are true, and include the poor memory score ,then colour me impressed!


-EDIT, just realized we're talking about passmark, which runs quite well on previous AMD uarch's, so really doesn't say a lot to be honest. Obviously benchmarks that already run favorably on construction cores aren't to going to see the full 40% uplift.. this being one of them.

if the ST performance compared to current Intel uarch's is representive of other workloads though, then it looks good.

 

[sm625]

Also there is a supposed Zen engineering sample picture going around with that same code, probably a Bristol Ridge AM4 part (thus the blacked out parts).

https://www.pic-upload.de/view-32649656/RyzenSampleRevision6.png.html

It is not bristol ridge. It is an 8 core Ryzen. Stop spreading FUD.

This is literally bulldozer all over again. It cant even beat a 4th generation i3.

 

[IllogicalGlory]

-

 

[.vodka]

@sm625, can you link to that particular Passmark CPU number result, #771904? I'm not finding it. Where did you get that picture from? What's the source?

Besides, Passmark reports CPUID string... that G3258 could very well be running at 5GHz, explaining how 6700k is getting beaten.

 

[lolfail9001]

Yet the SKUs listed lately all start with YD not ZD

1. There is a photo of the very same sample with literally same name written on it in one of these Zen threads. You have literally posted it here.
2. It is a qual sample from what i may guess, these were ZDs https://videocardz.com/65654/amd-ryzen-6-core-cpu-exists
3. I'll make it easy for you: there is no F4 revision of Bristol Ridge. So if you want to go into denial, your only option is to claim it was photoshopped photo.

This is literally bulldozer all over again. It cant even beat a 4th generation i3.

To start with: just post your source.

Real benchmarks: http://www.guru3d.com/news-story/fr...ng-sample-amd-ryzen-processor-benchmarks.html

If he posts a valid source, that link and his source are both real benchmarks.

 

[Doom2pro]

It cant even beat a 4th generation i3.

What's your point, in that benchmark apparently not even an i7 6700K @ 4Ghz can beat it either. This whole benchmark is FUD and entirely suspect.

 

[sm625]

@sm625, can you link to that particular Passmark CPU number result, #771904? I'm not finding it. Where did you get that picture from? What's the source?

Besides, Passmark reports CPUID string... that G3258 could very well be running at 5GHz, explaining how 6700k is getting beaten.

You need to download passmark and run it locally. It will be some time before it appears on their website.

My G3258 is running at 4.4 GHz.

 

[Rngwn]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

Hmm, a 3.8 GHz Ryzen losing out against the 3.6 GHz i3 on ST is a very telling about its IPC.

 

[revanchrist]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

Haswell @ 3.20Ghz beats Skylake @ 4.00Ghz in Single Thread performance. Sounds legit.

 

[Lepton87]

Just a few months ago I didn't think it would be possible for the Ryzen to beat my RIG at single-threaded performance but now I think it may actually happen. No one bothers to overclock the uncore when testing HEDT but increasing the uncore clock to 4.25GHz from 3GHz can sometimes yield as much as double digit gains with the average being over 5% so HW-E with the uncore overclocked by 40% matches BW IPC easily.

 

[IllogicalGlory]

It is not bristol ridge. It is an 8 core Ryzen. Stop spreading FUD.

This is literally bulldozer all over again. It cant even beat a 4th generation i3.

Assuming the benchmark is real and is a Ryzen chip at 3.4/3.8, it loses by 7% to a 3.9GHz (boost clock) Haswell chip while it's at presumably 3.8GHz itself, which makes it 4.5% behind Haswell IPC. Yup, sounds just as bad as Bulldozer, you're right! Again, assuming the results are even legitimate.

We still have these to consider: http://www.guru3d.com/news-story/fr...ng-sample-amd-ryzen-processor-benchmarks.html

And of course official numbers aren't here yet, so maybe we can ease back on the Faildozer talk? If Ryzen turns out to be really great, you and all the other people in here reaching for whatever negative news/rumors they can will be okay, I promise.

 

[Doom2pro]

1. There is a photo of the very same sample with literally same name written on it in one of these Zen threads. You have literally posted it here.

I also literally pointed that out a few minutes ago in this thread. That Engineering Sample photo only shows up recently, and is only referenced in posts about Ryzen pricing yet none of the SKUs start with ZD.

Very dubious.

 

[.vodka]

You need to download passmark and run it locally. It will be some time before it appears on their website.

My G3258 is running at 4.4 GHz.

Wait.

You are the source? Your 4.4GHz G3258 is beating a stock 6700k? You do realize you're linking a single threaded Passmark score where the 6700k can boost up to 4.2GHz? I'm not quite sure a 4.2GHz 6700k can be so far behind a 4.4GHz heavily cut down Haswell sample.

You have your hands on a Ryzen sample?

 

[lopri]

We finally have a Ryzen passmark baseline. Sorry, it is a turd.

Beaten to a pulp by my 3 year old $50 cpu.

The main passmark score is 15084, compared to 17683 of the 6900K.

The memory score is really bad too, at 1800.

You will have to excuse my incredulity.

I predict that the top tier 95W Zen CPU from the initial release group will have the following passmark scores:

13000 Multithread
1800 Single thread

This prediction is based on the fact that Nvidia engineering + TSMC 16nm appears to be more than 50% ahead of AMD engineering + GF14nm in terms of perf per watt. And Intel should at least be able to match Nvidia+TSMC, if not outright beat them.

Would you care to place a bet on what the passmark scores will be?

I do not know what it is that you have with that mediocre benchmark, but it seems obvious to me you become rather emotional when it comes to Zen products.