AMD responds to 1080p gaming tests on Ryzen. Supports ECC RAM. Win 10 SMT bug

[Valentyn]

5 minutes ago, PCGuy_5960 said:

Yes, but still the benchmarks that aren't wrong show that the 7700K beats the 1700 in games....

BTW, Intel does still win in his 1800X review (in games):

 

 

I'd certainly expect Kaby Lake at 5Ghz to win with a 1.1Ghz advantage.

It certainly beats my 4Ghz Haswell - E there as well.

[PCGuy_5960]

Just now, Valentyn said:

I'd certainly expect Kaby Lake at 5Ghz to win with a 1.1Ghz advantage.

It certainly beats my 4Ghz Haswell - E there as well.

Yes, but I am pretty sure that your 4GHz Haswell-E will beat the 4GHz Ryzen as well

[Drak3]

7 minutes ago, LAwLz said:

OK, so why not run the Ryzen system with 4 sticks of RAM? That way the RAM has not changed, the Intel system gets quad channel and the AMD system gets dual channel.

Using the 4 same sticks isn't a variable. Using them in different configurations is.

The whole point of what I'm saying in regards to dual vs quad channel is that ALL VARIABLES THAT CAN BE ELIMINATED, FOR SCEINTIFIC TESTING, SHOULD GET ELIMINATED.

 

 

@djdwosk97

But I'm not even saying that we should only do that. I'm saying that WE SHOULD TEST BOTH IDENTICAL CONFIGURATIONS AND CONFIGURATIONS THAT ARE MORE LIKELY TO BE SEEN IN THE REAL WORLD, TO COVER ALL BASES.

 

And I shouldn't need to say it again. If I do, I'm going to say it to a brick wall. Evidence would lead to the conclusion - a brick wall is significantly fucking smarter.

[79wjd]

4 minutes ago, Drak3 said:

 

@djdwosk97

But I'm not even saying that we should only do that. I'm saying that WE SHOULD TEST BOTH IDENTICAL CONFIGURATIONS AND CONFIGURATIONS THAT ARE MORE LIKELY TO BE SEEN IN THE REAL WORLD, TO COVER ALL BASES.

 

And I shouldn't need to say it again. If I do, I'm going to say it to a brick wall. Evidence would lead to the conclusion - a brick wall is significantly fucking smarter.

And how many people do you think are going to run a 6900k with their memory in dual channel? .0001%? That's definitely a worthwhile percentage to gimp X99 in order to make Ryzen look better. While we're at it, let's test to see how both of them run in a 60ºc oven, and in space, and at 1.1ghz, and.....

[Valentyn]

5 minutes ago, PCGuy_5960 said:

Yes, but I am pretty sure that your 4GHz Haswell-E will beat the 4GHz Ryzen as well

 

I'd love to actually test it myself, sadly the NASDAQ took a nasty hit the last two days, so I shan't indulge in a complete new system just yet, keeping an eye out for second hand 5960X and 6900K now

Seems the chance of an Interest Hike is causing many short term traders to cut profits and run.

NVIDIA have been going down since they announced the GTX 1080ti, and AMD at Capsaicin; although they took a further nasty hit today after the reviews showing more than questionable gaming results.

[PCGuy_5960]

3 minutes ago, Valentyn said:

I'd love to actually test it myself, sadly the NASDAQ took a nasty hit the last two days, so I shan't indulge in a complete new system just yet, keeping an eye out for second hand 5960X and 6900K now

Sadly though. I think that performance in games won't improve by more than 5-10%...

[Drak3]

14 minutes ago, djdwosk97 said:

And how many people do you think are going to run a 6900k with their memory in dual channel? .0001%? That's definitely worthwhile testing for. 

@LAwLz

So, get this, told the brick wall, and it didn't spit out the same bullshit as you are whilst completely ignoring what I'm saying.

[LAwLz]

1 minute ago, Drak3 said:

Using the 4 same sticks isn't a variable. Using them in different configurations is.

The whole point of what I'm saying in regards to dual vs quad channel is that ALL VARIABLES THAT CAN BE ELIMINATED, FOR SCEINTIFIC TESTING, SHOULD GET ELIMINATED.

 

 

@djdwosk97

But I'm not even saying that we should only do that. I'm saying that WE SHOULD TEST BOTH IDENTICAL CONFIGURATIONS AND CONFIGURATIONS THAT ARE MORE LIKELY TO BE SEEN IN THE REAL WORLD, TO COVER ALL BASES.

 

And I shouldn't need to say it again. If I do, I'm going to say it to a brick wall. Evidence would lead to the conclusion - a brick wall is significantly fucking smarter.

So what you are saying is that you should in fact kneecap products which has features the other one does not, because that is more fair?

So if we were to compare an AMD graphics card vs an Nvidia one we should deliberately disable everything related to a-sync compute? Because that would eliminate variables.

 

Also, don't you think benchmarks should test what consumers actually end up with? I mean, by your logic it would be completely fair for Intel to say "we should disable Hyperthreading and SMT for these tests because that eliminates variables". Or how about disabling cores on the 1800X when comparing it to the 7700K? Core count is a "variable" just as much as quad channel support is, right? So why not "try and remove variables"?

Oh right, it's only fair to "remove variables" when it benefits AMD, but not the other way around, right?

 

Your entire posts basically boils down to "baww stop using things which makes Intel's platform superior to AMD's! We should kneecap Intel so that the AMD chip appears to be more competitive than it is!". Hey, Intel doesn't have XFR. Should that be disabled too? Oh right, it shouldn't because "it's part of the chip", even though quad channel memory is also part of the chip (it's a feature of the IMC which is built into the CPU die).

[Valentyn]

6 minutes ago, PCGuy_5960 said:

Sadly though. I think that performance in games won't improve by more than 5-10%...

 

Can only hope, but I think there's room for improvement, atleast 5% which puts the onpar with Haswell-E which is what many of us expected. 

 

Fingers crossed because I'd love some real good competition. At least it's not another Bulldozer where it's over 25% slower than Sandy Bridge on its launch day.

 

[PCGuy_5960]

Just now, Valentyn said:

Can only hope, but I think there's room for improvement, atleast 5% which puts the onpar with Haswell-E which is what many of us expected. 

BTW... In the video that you linked, AMD talked about MSI boards having issues.... ASUS boards didn't have an issue that affected gaming performance AFAIK...

 

[Valentyn]

1 minute ago, PCGuy_5960 said:

BTW... In the video that you linked, AMD talked about MSI boards having issues.... ASUS boards didn't have an issue that affected gaming performance AFAIK...

 

 

They mention anomalous performance on ASUS, and Nexus mentioned MSI does as well. I think I linked too far in, go back a few minutes.

 

 

[PCGuy_5960]

4 minutes ago, Valentyn said:

 

They mention anomalous performance on ASUS, and Nexus mentioned MSI does as well. I think I linked too far in, go back a few minutes.

Oh... Yeah... Hardware Unboxed did say that they had such issues with the ASUS board, but the AsRock one had no such issues... 

(BTW, AMD said that they noticed enormous performance improvements with the new BIOS on the MSI board, but they didn't say something like that for the ASUS one)

[Valentyn]

Just now, PCGuy_5960 said:

Oh... Yeah... Hardware Unboxed did say that they had such issues with the ASUS board, but the AsRock one had no such issues... 

(BTW, AMD said that they noticed enormous performance improvements with the new BIOS on the MSI board, but they didn't say something like that for the ASUS one)

Yup, I don't think Asus delivered a better BIOS in time for the rushed launch.
Golem.de saw an average of 17% improvement on MSI with the newer BIOS. 

Interestingly ASUS were rather bragging early on about better memory speed support as well, yet they couldn't really deliver there also. Hope that's all sorted soon, although looking UK retailers there's a massive motherboard shortage compared to CPUs as well.

 

[PCGuy_5960]

3 minutes ago, Valentyn said:

Golem.de saw an average of 17% improvement on MSI with the newer BIOS. 

That's INSANE... MSI fucked up big time

4 minutes ago, Valentyn said:

Yup, I don't think Asus delivered a better BIOS in time for the rushed launch.

Found another review with a Gigabyte board:

And the 6900K is much faster in this one as well.... (Gaming of course)

[Valentyn]

6 minutes ago, PCGuy_5960 said:

That's INSANE... MSI fucked up big time

Found another review with a Gigabyte board:

And the 6900K is much faster in this one as well.... (Gaming of course)

 

Yeah this is all bizarre, I wonder how much is down to shitty drivers/software, and BIOS; with the rest laying on Ryzen.
It's certainly wise for people to hold off a minimum of a month before buying.

 

Even my X99 Deluxe motherboard had lots of issues on launch and shortly after. Memory and XMP speeds also being one, wish I waited longer there before buying; as I also got ripped off hard on RAM. ( It cost more for my ram, than 5820K and Mobo combined )
 

[PCGuy_5960]

Just now, Valentyn said:

Even my X99 Deluxe motherboard had lots of issues on launch and shortly after. Memory and XMP speeds also being one, wish I waited longer there before buying; as I also got ripped off hard on RAM. ( It cost more for my ram, than 5820K and Mobo combined )

I got my 5820K in June, it was dirt cheap 

1 minute ago, Valentyn said:

Yeah this is all bizarre, I wonder how much is down to shitty drivers/software, and BIOS; with the rest laying on Ryzen.

However... Gigabyte did an amazing job from what we can see... Performance is definitely better with the AORUS motherboard, but still not up to par with Broadwell-E..

[Mira Yurizaki]

52 minutes ago, Valentyn said:

NVIDIA have been going down since they announced the GTX 1080ti, and AMD at Capsaicin; although they took a further nasty hit today after the reviews showing more than questionable gaming results.

NVIDIA's going down because some financial analyst said to sell after their little bubble. And this was at least days before the GTX 1080 Ti announcement.

[Valentyn]

4 minutes ago, M.Yurizaki said:

NVIDIA's going down because some financial analyst said to sell after their little bubble. And this was at least days before the GTX 1080 Ti announcement.

Analysts have been saying NVIDIA is overvalued for weeks now. The entire NASDAQ is down because of impending interest hikes.

 

http://www.cnbc.com/2017/03/01/chance-of-a-march-rate-hike-doubled-cme.html

[Valentyn]

22 minutes ago, PCGuy_5960 said:

I got my 5820K in June, it was dirt cheap 

However... Gigabyte did an amazing job from what we can see... Performance is definitely better with the AORUS motherboard, but still not up to par with Broadwell-E..

I got my setup shortly after launch...DDR4 prices on launch..*shudder*.
UI couldn't get RAM anywhere, and ended up just buying the only sticks left in stock.

 

It took Asus nearly 3 months before I could even run my RAM at 2400Mhz without issues. DDR4 launch was messy as well.

[PCGuy_5960]

1 minute ago, Valentyn said:

 

HOLY SHIT

[Valentyn]

33 minutes ago, PCGuy_5960 said:

HOLY SHIT

Early adopters always suffer, if not cost, in hardware and software niggles sadly.

 

Hoping AMD irons it out, as I would like to either get one of their later Naples CPUs for a workstation, or a Skylake-E one; at a decent price that is. I don't like the idea of the 8core Intel being  a grand, but also like my gaming and want it to perform there.

[spartaman64]

intel at least seem to think that amd can resolve these problems considering their price cuts

[zMeul]

1 minute ago, spartaman64 said:

intel at least seem to think that amd can resolve these problems considering their price cuts

mate, what price cuts?

[alextulu]

On 3/2/2017 at 9:05 PM, zMeul said:

@Curious Pineapple

 

I find this really fuckign funny since HPET was co-developed by AMD and MS starting 2005

now they're blaming HPET and BIOS not being on par .. what the fuck are they doing over there?! it's like they hired a guy to design the CPU, fired him, sent the designs to the fab, got the product and now they don't know what to do with it 

This article explains it better: https://www.legitreviews.com/amd-ryzen-7-1800x-1700x-and-1700-processor-review_191753/15

Quote

They also have a Windows Driver coming in approximately one month that will help performance as the Windows High Precision Event Timer (HPET) isn’t playing nice with the SenseMI sensors that poll the CPU status every millisecond.

There's a conflict between HPET and SenseMI.

Quote

Make sure the system has Windows High Precision Event Timer (HPET) disabled. HPET can often be disabled in the BIOS. Alternatively: from Windows, open an administrative command shell and type:  bcdedit /deletevalue useplatformclock  – this can improve performance by 5-8%.

 

 

About Ryzen's memory issues: http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/74814-amd-ryzen-7-1800x-performance-review-5.html

 

 

Quote

How do you avoid dual rank memory kit? Well, first and foremost, most memory manufacturers will be launching AMD-specific models immediately.

Quote

From a consumer point-of-view, there is no reason to buy dual rank modules, unless you need 16GB modules, in which case you currently have no other option

 

[Dabombinable]

13 minutes ago, alextulu said:

This article explains it better: https://www.legitreviews.com/amd-ryzen-7-1800x-1700x-and-1700-processor-review_191753/15

There's a conflict between HPET and SenseMI.

About Ryzen's memory issues: http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/74814-amd-ryzen-7-1800x-performance-review-5.html

 

 

 

Dual rank=chips on both sides of a stick.