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

[lolfail9001]

Dolphin doesn't have any MOVBE instructions in it, and the only intruction belonging to BMI(1) is TZCNT.

No 256-bit instructions either.

This is for the public 5.0 binary available at the site.

Then why the hell is Haswell like 60-70% faster than Ivy? What is this sorcery?

Yes they can, the other cpus have recorded turbos and the 1800X like the other had 1700X which could be firmware disabled. Passmark has a built in CPUID which detects turbo on or off through BIOS. It was a 3.6 baseline validation score.

Consider for a second that CPU-Z screen of recent could not detect multiplier range on Zen either and just reported 34.5x multiplier on 37/33 SKU. Turbo is hidden, deal with it.

 

[unseenmorbidity]

 

[Doom2pro]

What's up with that SSE score? Looks like AMD flipped Intel the bird!

 

[OrangeKhrush]

What's up with that SSE score? Looks like AMD flipped Intel the bird!

It is good at very obvious things, but still nothing new, I think we are getting more and more of the same thing

 

[Shivansps]

Actually, from that i could extrapolate that the old Ryzen passmark result was actually running with turbo enabled to 3.8Ghz and this one indeed it seems turbo disabled and 3.6ghz, both cases seems to indicate below than Haswell IPC when compared to a 3.6 and 3.8 Haswell I3s, and more in line with the what AMD promised before all the hype.

 

[OrangeKhrush]

Actually, from that i could extrapolate that the old Ryzen passmark result was actually running with turbo enabled to 3.8Ghz and this one indeed it seems turbo disabled and 3.6ghz, both cases seems to indicate below than Haswell IPC when compared to a 3.6 Ghz Haswell I3, and more in line with the what AMD promised before all the hype.

I think the other one was 3.7, as going up from 1980 to 3.7 is about 2030 either way yeah seems like the otherone was hidden this one is off or turbo is just not working well at all. IPC about Ivyish which is barely acceptable levels

 

[unseenmorbidity]

Actually, from that i could extrapolate that the old Ryzen passmark result was actually running with turbo enabled to 3.8Ghz and this one indeed it seems turbo disabled and 3.6ghz, both cases seems to indicate below than Haswell IPC when compared to a 3.6 and 3.8 Haswell I3s, and more in line with the what AMD promised before all the hype.

The single threaded benchmark shows that the performance is basically identical to broadwell-e clock for clock. It has a sizable lead over the 5960x at nearly the same clocks.

 

[inf64]

Actually, from that i could extrapolate that the old Ryzen passmark result was actually running with turbo enabled to 3.8Ghz and this one indeed it seems turbo disabled and 3.6ghz, both cases seems to indicate below than Haswell IPC when compared to a 3.6 and 3.8 Haswell I3s, and more in line with the what AMD promised before all the hype.

What is below Haswell level? I usually consider 2-3% deviation to be within margin of error

 

[vissarix]

I3 4360 3.7ghz CB 149
Ryzen r5 1600x 3.7ghz CB 146

http://cdn.wccftech.com/wp-content/...yzen-5-1600X-Cinebench-R15-Benchmark-Leak.png
So there you have it...lower IPC compared to haswell...obviously way higher multiscore and lower price will help Ryzen overcome this...

 

[Shivansps]

I think the other one was 3.7, as going up from 1980 to 3.7 is about 2030 either way yeah seems like the otherone was hidden this one is off or turbo is just not working well at all. IPC about Ivyish which is barely acceptable levels

It was 38/34 i still had it in my passmark suite. Ill post a image.

 

[inf64]

I3 4360 3.7ghz CB 149
Ryzen r5 1600x 3.7ghz CB 146

So there you have it...lower IPC compared to haswell...obviously way higher multiscore and lower price will help Ryzen overcome this...

2 percent lower? Wow...That is within margin of error.

 

[Shivansps]

What is below Haswell level? I usually consider 2-3% deviation to be within margin of error

This is what happen when people take passmark results with ENG samples and show them as clickbait articles, we cant be really sure of anything, but i a take passmark as a aceptable bench, thats the result unfortunatelly.

 

[OrangeKhrush]

It was 38/34 i still had it in my passmark suite. Ill post a image.

it works out to.

 

[CatMerc]

I don't know what WCCFTech is smoking, I just did the SSE test on my 6600K @ 4.2GHz, and got 474 points.
Apparently I have a magical CPU that gets 5.2x the perfomance of a CPU with half its cores

 

[OrangeKhrush]

I don't know what WCCFTech is smoking, I just did the SSE test on my 6600K @ 4.2GHz, and got 474 points.
Apparently I have a magical CPU that gets 5.2x the perfomance of a CPU with half its cores

or passmark is very loopy

 

[inf64]

I don't know what WCCFTech is smoking, I just did the SSE test on my 6600K @ 4.2GHz, and got 474 points.
Apparently I have a magical CPU that gets 5.2x the perfomance of a CPU with half its cores

You need to take a closer look at what they (wccwhatever) used for results. Notice (PT8) next to some of the scores? Yes , that means it is taken from an older(different) version of Passmark and they are not directly comparable. Bad journalism unfortunately.

 

[CatMerc]

Could someone with pre-Skylake CPU test it? It appears my CPU is 99% percentile, which it's not in any other test.
Maybe Skylake did something that Broadwell didn't in that regard.

You need to take a closer look at what they (wccwhatever) used for results. Notice (PT8) next to some of the scores? Yes , that means it is taken from an older(different) version of Passmark and they are not directly comparable. Bad journalism unfortunately.

Oh, that would explain it lol

 

[OrangeKhrush]

You need to take a closer look at what they (wccwhatever) used for results. Notice (PT8) next to some of the scores? Yes , that means it is taken from an older(different) version of Passmark and they are not directly comparable. Bad journalism unfortunately.

sounds like WCCFT

 

[lolfail9001]

I don't know what WCCFTech is smoking, I just did the SSE test on my 6600K @ 4.2GHz, and got 474 points.

WCCFTech are smoking clickbait, it is super effective.

 

[CentroX]

I3 4360 3.7ghz CB 149
Ryzen r5 1600x 3.7ghz CB 146

http://cdn.wccftech.com/wp-content/...yzen-5-1600X-Cinebench-R15-Benchmark-Leak.png
So there you have it...lower IPC compared to haswell...obviously way higher multiscore and lower price will help Ryzen overcome this...

Ipc is within 5% of skylake not sure what you are smoking.

 

[OrangeKhrush]

WCCFTech are smoking clickbait, it is super effective.

Net traffic is good money.

 

[CentroX]

Ryzen is a hybrid of 7700k and 6900k which is extremely impressive. Not sure why everyone is doom and gloom again?

 

[lolfail9001]

Ryzen is a hybrid of 7700k and 6900k which is extremely impressive. Not sure why everyone is doom and gloom again?

Ryzen looks like cheap Haswell-E, not anything even closely resembling 7700k.

Nothing wrong with just cheap Haswell-E in my book though.

 

[OrangeKhrush]

It is taking features on intel X and merging with socket 11XX with the obvious trade offs trying to fit all that onto a die the size of a comercial CPU. It doesn't have nearly as much resources as Xeon lite X99 chips but it has a lot compared with the mainstream i7's. Some other sacrifices were made in balancing peformance to cost. It should be fast though.

 

[CentroX]

Ryzen looks like cheap Haswell-E, not anything even closely resembling 7700k.

Nothing wrong with just cheap Haswell-E in my book though.

You realize that single thread performance of ryzen is very close 7700k? You are fail here.