AM4 mobo's BIOS performance bias?!

[Fetzie]

Just now, zMeul said:

except you know ...

• doesn't increase score for single card benches
• ^ same for dual card SLI
• nVidia told people even before the cards were out

The top 16 benchmarks on http://www.3dmark.com/hall-of-fame-2/timespy+3dmark+score+performance+preset/version+1.0 are with four TitanX Pascal.

 

This is no different, except it's for epeen on a different benchmark.

[zMeul]

Just now, RevoltTrain said:

Pretty sure Intel does the same thing for pre release type stuff

care to source that? or do you like to throw allegations around like Semi Accurate's Demerjian 

[NumLock21]

I don't see the problem, if the reviewer are honest with their results. They can show us both the standard settings and custom settings. The only problem are individuals who are having a e-peen competition, where one with this secret setting does not tell anyone about it.

[zMeul]

1 minute ago, Fetzie said:

The top 16 benchmarks on http://www.3dmark.com/hall-of-fame-2/timespy+3dmark+score+performance+preset/version+1.0 are with four TitanX Pascal.

 

This is no different, except it's for epeen on a different benchmark.

people don't buy cards based on quad SLI benchmarks 

[RevoltTrain]

Just now, zMeul said:

care to source that? or do you like to throw allegations around like Semi Accurate's Demerjian 

It's been posted before, this is fairly standard practice. Don't believe me, doesn't bother me any. 

[zMeul]

Just now, NumLock21 said:

I don't see the problem, if the reviewer are honest with their results. They can show us both the standard settings and custom settings. The only problem are individuals who are having a e-peen competition, where one with this secret setting does not tell anyone about it.

it would be likely be missed, it's enabled by default and if the NDA/review window is short reviewers will hurry to get it done a.s.a.p.

[zMeul]

1 minute ago, RevoltTrain said:

It's been posted before, this is fairly standard practice. Don't believe me, doesn't bother me any. 

prof is in the pudding, where is the puddin' 

[NumLock21]

Just now, zMeul said:

it would be likely be missed, it's enabled by default and if the NDA/review window is short reviewers will hurry to get it done a.s.a.p.

Motherboard manufactures will probably inform the reviewer about this new feature.

[zMeul]

Just now, NumLock21 said:

Motherboard manufactures will probably inform the reviewer about this new feature.

like they did with the video cards .. right 

[Fetzie]

 

1 minute ago, zMeul said:

prof is in the pudding, where is the puddin' 

 

NDA on reviews lifts in a few hours. Then we'll see the pudding (whatever it's flavour).

[zMeul]

Just now, Fetzie said:

NDA on reviews lifts in a few hours. Then we'll see the pudding.

it will be a shit show

and if they didn't knew about it, they'll retest everything .. if they value their integrity

[goodtofufriday]

16 minutes ago, zMeul said:

special designed BIOS to artificially increase results in select benchmarks

With what evidence? A single person showing a bios of a platform that no one has? 

I typically respect your opinions but youre hella jumping the gun here. 

[zMeul]

17 minutes ago, goodtofufriday said:

With what evidence? A single person showing a bios of a platform that no one has? 

I typically respect your opinions but youre hella jumping the gun here. 

apparently it's available on ASUS site if you search hard enough https://www.asus.com/fr/Motherboards/ROG-CROSSHAIR-VI-HERO/HelpDesk_Download/

 

 

how you skew perception of a product before it launches? you do "select" benchmarks in particular ways - and this is exactly what this looks like

[MakeAMDGreatAgain]

4 minutes ago, Fetzie said:

 

 

NDA on reviews lifts in a few hours. Then we'll see the pudding (whatever it's flavour).

Is there a countdown page? I'm hyped

[Notional]

2 minutes ago, MakeAMDGreatAgain said:

Is there a countdown page? I'm hyped

https://countingdownto.com/countdown/amd-ryzen-release-countdown-clock

[MakeAMDGreatAgain]

Just now, Notional said:

https://countingdownto.com/countdown/amd-ryzen-release-countdown-clock

He said a few hours.. wtf! haha, anyways.. Thanks got it bookmarked

[goodtofufriday]

4 minutes ago, zMeul said:

apparently it's available on ASUS site if you search hard enough fuck french, this was not true, I understood wrong 

 

how you skew perception of a product before it launches? you do "select" benchmarks in particular ways - and this is exactly what this looks like

This os what i found. Its a bios feature on asus server platforms with the following description 

 

"Energy Performance Bias setting [Balanced Performance]

Allows you to set the Energy Performance Bias which overrides the OS setting.

Configuration options: [Performance] [Balanced Performance] [Balanced Power]

[Power] "

 

RS400-E8-PS2 is the server board. 

[zMeul]

1 minute ago, goodtofufriday said:

This os what i found. Its a bios feature on asus server platforms with the following description 

 

"Energy Performance Bias setting [Balanced Performance]

Allows you to set the Energy Performance Bias which overrides the OS setting.

Configuration options: [Performance] [Balanced Performance] [Balanced Power]

[Power] "

 

RS400-E8-PS2 is the server board. 

what are "we" talking about?!

 

I already discussed this with @Enderman

what is shown in the ASUS BIOS and what the settings does on that SuperMicro board is apples to oranges difference

[Guest Kloaked]

I just want to preface that I cannot provide proof for this, but anyways:

 

I used to have an Asus Maximus V Gene (Z77) for my 2500k that is now on the shelf. I ran benchmarks for it at the usual 4.5Ghz. I also had two GTX 770s I was using at the time, but at some point one of the PCI slots stopped working so I couldn't use SLI. I had do get another motherboard and this was almost when Haswell launched, so naturally there weren't any "new" boards that I could buy. The only thing I could find that was halfway decent was an ASRock Z77 Extreme4.

 

I ran benchmarks once more, at 4.5Ghz and low and behold, I had a good chunk lower score than I did before I swapped mobos.

 

So take that for what it's worth: mostly a grain of salt, but also some information that could tell more to the story (maybe the motherboard you buy does matter, for example).

[Fetzie]

 

 

Linus living up to his reputation (it isn't off-topic, shows that LTT has Asus motherboards so should have access to this feature)

[Dabombinable]

15 minutes ago, zMeul said:

because that's exactly what happened when GTX10 series AIB cards launched - MSI and ASUS sent reviewers special cards clocked higher than retail ones

let's not pretend it's not an issue

Gigabyte is as bad as well. They gave reviewers amongst the cream of the crop of their GTX 970 G1 Gaming range, and sold the rest pretty much only binned for being capable of 1329MHz-with no overclocking headroom at all for some despite what the box says about them being binned for overclocking.

[Notional]

8 minutes ago, MakeAMDGreatAgain said:

He said a few hours.. wtf! haha, anyways.. Thanks got it bookmarked

NDA press date is 2nd March 9AM CT (slide 2-51). Make sure the countdown is that time, the guy who made it, might hade derped.

[zMeul]

1 minute ago, Dabombinable said:

Gigabyte is as bad as well. They gave reviewers amongst the cream of the crop of their GTX 970 G1 Gaming range, and sold the rest pretty much only binned for being capable of 1329MHz-with no overclocking headroom at all for some despite what the box says about them being binned for overclocking.

I dunno about that

but I provided a very recent example of what happened less than 1y ago

[MakeAMDGreatAgain]

2 minutes ago, Notional said:

NDA press date is 2nd March 9AM CT (slide 2-51). Make sure the countdown is that time, the guy who made it, might hade derped.

Hmm no way to change it on that page.

[goodtofufriday]

8 minutes ago, zMeul said:

what are "we" talking about?!

 

I already discussed this with @Enderman

what is shown in the ASUS BIOS and what the settings does on that SuperMicro board is apples to oranges difference

@_@ confused. 

 

Im saying that the performance bias seen in that tweet is likely the same as in the server board. 

 

Whats this about super micro?