Regarding the RX 480 AOTS benchmark at Computex

[Briggsy]

1 hour ago, suicidalfranco said:

LTT is the only forum i visit, don't even know what's reddit ¯\_(ツ)_/¯

Reddit is the epitomy of an echo chamber, where people upvote posts that confirm their own bias, and downvote posts that challenge their bias. The only way to get a less than favorable point across on reddit is to say something agreeable at the start of the post that people don't want to downvote.

 

Reddit is great for looking at silly cat pictures and for people posting great short stories or funny news articles, but for tech stuff it can be a seething cesspool of fanboyism and half-truths.

[zMeul]

1 minute ago, Ryan_Vickers said:

well there was the whole 364 fiasco... that's what I'm thinking

what exactly are you talking about?

[vanished]

2 minutes ago, zMeul said:

what exactly are you talking about?

https://linustechtips.com/main/topic/585316-update-36510-is-also-broken-urgent-nvidias-364xx-driver-is-broken

 

[zMeul]

2 minutes ago, Ryan_Vickers said:

https://linustechtips.com/main/topic/585316-update-36510-is-also-broken-urgent-nvidias-364xx-driver-is-broken

 

sorry, I don't use F@H or similar; didn't had issues with that driver

I do however have issues with the current set, 368.22 - Chrome freezes PC; I'm using 365.19

[vanished]

1 minute ago, zMeul said:

sorry, I don't use F@H or similar; didn't had issues with that driver

I do however have issues with the current set, 368.22 - Chrome freezes PC; I'm using 365.19

well there you go  a driver issue

[zMeul]

1 minute ago, Ryan_Vickers said:

well there you go

still don't understand where you getting at 

[Trixanity]

16 minutes ago, zMeul said:

what are you talking about?

 

AMD used pre-release drivers instead of the publicly available 368.25 WHQL, since May 26th

It's generally not a good idea to change things less than a week from deadline. While it's unlikely things would go wrong, there is a chance. The new driver could introduce new bugs. They played it safe by sticking to what they've tested in their labs and that was brought onto the stage.

[vanished]

3 minutes ago, zMeul said:

still don't understand where you getting at 

You said you didn't have any issues with the one I mentioned or know of it, but you're having your own issues with a different version.  Either way, they seem to really be slipping up lately.  I've been using Nvidia since 2002 and never had a driver issue I can remember, and even here on the forum they are rare (or have been), yet there's all these stories of late....

[zMeul]

4 minutes ago, Trixanity said:

It's generally not a good idea to change things less than five days from deadline. While it's unlikely things would go wrong, there is a chance. The new driver could introduce new bugs. They played it safe by sticking to what they've tested in their labs and that was brought onto the stage.

no, that was the wrong thing to do since AMD used a pre-release driver

they could've at least checked the new driver if it performed differently - well ... it did; and the asshats from AMD's PR went with the most plausible reddit created excuse instead of doing some research on their own

and people wonder why I bash AMD so "violently" ... because they do shit like this

[zMeul]

5 minutes ago, Ryan_Vickers said:

You said you didn't have any issues with the one I mentioned or know of it, but you're having your own issues with a different version.  Either way, they seem to really be slipping up lately.  I've been using Nvidia since 2002 and never had a driver issue I can remember, and even here on the forum they are rare (or have been), yet there's all these stories of late....

ATi had driver issues, AMD has driver issues, nVidia has driver issues - it's an ongoing shit show

[Trixanity]

9 minutes ago, zMeul said:

no, that was the wrong thing to do since AMD used a pre-release driver

they could've at least checked the new driver if it performed differently - well ... it did; and the asshats from AMD's PR went with the most plausible reddit created excuse instead of doing some research on their own

and people wonder why I bash AMD so "violently" ... because they do shit like this

Yes, AMD used a pre-release driver. For both sides. Contrary to what you might think: you don't change things before a big presentation mere days ahead unless you have a big problem which you need to solve (eg malfunctions). There are countless examples of presentations going wrong because things aren't tested properly. Nvidia or Intel would have done the exact same thing because it would be foolish to bring an unknown quantity to such a big audience. It's fine for small, closed door prototype reveals but for something like this? No way. That's begging for trouble.

 

And what exactly was the excuse? It did render incorrectly. It perhaps gained a little bit performance by doing so. Perhaps it didn't. Perhaps it lost a little. The setup and settings used have been well documented by now. And everything will be independently verifiable soon.

 

Your excuse for ranting is weak at best and worst of all: you know it.

[zMeul]

11 minutes ago, Trixanity said:

Yes, AMD used a pre-release driver. For both sides. Contrary to what you might think: you don't change things before a big presentation mere days ahead unless you have a big problem which you need to solve (eg malfunctions).

sorry, but that's not how it works

AMD knew very well they used a pre-release driver - and that should've been disclosed 1st hand; then checked against the WHQL driver as they had 4-5 days to do it and re-do it

 

that's a shitty excuse for major amateur hour in AMD's camp -  at the time AMD had their conference, nVidia had released the GTX1080 for ~5 days

they wanted to make nVidia look bad with that bullcrap presentation, karma is a bitch 

[Trixanity]

1 minute ago, zMeul said:

sorry, but that's not how it works

AMD knew very well they used a pre-release driver - and that should've been disclosed 1st hand; then checked against the WHQL driver as they had 4-5 days to do it and re-do it

 

that's a shitty excuse for major amateur hour in AMD's camp

Oh but it is how it works. Don't know about down in the village, but in a cooperate scenario: you don't want things going wrong on stage. AMD even stated they tested it 10 times just before going on stage. That's how important it is to be absolutely sure that everything is in order.

While they might have been able to slip in a new driver a few days before and run benchmarks on it. What if the machine crashed on stage? Or the FPS dropped by 5? Or it increased by 5? What if the rendering was the same? (has it actually been absolutely confirmed it has? Devs seem to think so) It would have changed their whole presentation. That's why you don't mess around.

 

AMD didn't list driver versions for either, besides it seemed to have benefited Nvidia. Best case scenario: extra FPS and apparently the picture looked prettier, which was funnily enough why it was revealed that Nvidia's previous driver was buggy.

 

So thank the likes of you for that. If you kept quiet, Nvidia would have come out looking good but some ranting Nvidia fanboys had to ruin it all. If it weren't for those meddling kids, amirite?

[79wjd]

30 minutes ago, Trixanity said:

Oh but it is how it works. Don't know about down in the village, but in a cooperate scenario: you don't want things going wrong on stage. AMD even stated they tested it 10 times just before going on stage. That's how important it is to be absolutely sure that everything is in order.

While they might have been able to slip in a new driver a few days before and run benchmarks on it. What if the machine crashed on stage? Or the FPS dropped by 5? Or it increased by 5? What if the rendering was the same? (has it actually been absolutely confirmed it has? Devs seem to think so) It would have changed their whole presentation. That's why you don't mess around.

 

AMD didn't list driver versions for either, besides it seemed to have benefited Nvidia. Best case scenario: extra FPS and apparently the picture looked prettier, which was funnily enough why it was revealed that Nvidia's previous driver was buggy.

 

So thank the likes of you for that. If you kept quiet, Nvidia would have come out looking good but some ranting Nvidia fanboys had to ruin it all. If it weren't for those meddling kids, amirite?

Yes, what if the 1080 crashed or it produced lower FPS...that would just be awful for AMD's marketing. The only way it could play out badly for AMD is if the 480 based system crashed or if the 1080 had a higher frame rate. 

[Big-G]

5 hours ago, Notional said:

Officially confirmed that NVidia fracked up their shitty drivers yet again:

 

 

It's called generational driver maturity... new uArch, early drivers...

 

3 hours ago, Notional said:

And people blaming AMD for the result 

More like AMD fanboys claiming hardware issues... jumping to conclusions...

 

2 hours ago, zMeul said:

what are you talking about?

 

AMD used pre-release drivers instead of the publicly available 368.25 WHQL, since May 26th

Agreed, deliberate AMD gimping using press drivers and not public drivers

 

1 hour ago, zMeul said:

ATi had driver issues, AMD has driver issues, nVidia has driver issues - it's an ongoing shit show

Give that man a wisky!

 

 

[ace_cheaply]

17 minutes ago, GidonsClaw said:

 

Agreed, deliberate AMD gimping using press drivers and not public drivers

 

Give that man a wisky!

 

 

This is deliberately ignoring the fact that the supposed gimping made the scene easier for the Nvidia gpu to render.

[Big-G]

42 minutes ago, ace_cheaply said:

This is deliberately ignoring the fact that the supposed gimping made the scene easier for the Nvidia gpu to render.

That's still not Nvidia's fault in this respect... AMD chose a driver that had the issue, and seems is not an issue in a later version. So again... how is a bugged Demo, that caused less processing in this instance and is since fixed later drivers Nvidia's gimping/fault?

[ace_cheaply]

21 minutes ago, GidonsClaw said:

That's still not Nvidia's fault in this respect... AMD chose a driver that had the issue, and seems is not an issue in a later version. So again... how is a bugged Demo, that caused less processing in this instance and is since fixed later drivers Nvidia's gimping/fault?

My comment was about you labeling it as "gimping" the 1080when it made it easier for it to render.  Nowhere did I say it was Nvidia's fault. 

[xAcid9]

59 minutes ago, GidonsClaw said:

That's still not Nvidia's fault in this respect... AMD chose a driver that had the issue, and seems is not an issue in a later version. So again... how is a bugged Demo, that caused less processing in this instance and is since fixed later drivers Nvidia's gimping/fault?

Yeah, definitely AMD's fault for not reading the changelog in latest driver.

[hex4]

I have plenty of issues on NV drivers and i'm using a VERY WELL matured GPU.

 

I will be going RX 480 myself.

[zMeul]

2 hours ago, ace_cheaply said:

This is deliberately ignoring the fact that the supposed gimping made the scene easier for the Nvidia gpu to render.

by AMD on AMD's presentation .. gimme a break!

AMD is the only one here that compares their products against competition - and they don't even do it properly

[ace_cheaply]

4 minutes ago, zMeul said:

by AMD on AMD's presentation .. gimme a break!

AMD is the only one here that compares their products against competition - and they don't even do it properly

Ok, take a break. 

[shdowhunt60]

9 hours ago, zMeul said:

what are you talking about?

 

AMD used pre-release drivers instead of the publicly available 368.25 WHQL, since May 26th

Okay? And they likely had that presentation completed before then.

[zMeul]

11 minutes ago, shdowhunt60 said:

Okay? And they likely had that presentation completed before then.

they don't bother to check the new driver that was available 5 days prior

then accuse nVidia of shit!!? again ...

 

how fucking stupid are they?! the entire AMD's PR team should get fired for incompetence

[shdowhunt60]

2 minutes ago, zMeul said:

they don't bother to check the new driver that was available 5 days prior

then accuse nVidia of shit!!? again ...

 

how fucking stupid are they?! the entire AMD's PR team should get fired for incompetence

Someone's mad.