AMD IQ Vs Nvidia IQ - shenanigans or something else?

tayto_0 · 3 Jul 2015 at 22:03

It seems that no matter what Greg does he can't win around here. From glancing at the other thread it seems all he was doing was comparing the TX with the FURY X, it wasn't even him who picked up on the difference in IQ.

Personally i don't think there's anything shady going on here and there's most likely a simple explanation for it.

Gregster · 3 Jul 2015 at 22:13

Kaapstad said:
I don't think you have missed anything out, I also think you have been very professional in your review.

More of the same please.

Thanks Kaap

tayto_0 said:
It seems that no matter what Greg does he can't win around here. From glancing at the other thread it seems all he was doing was comparing the TX with the FURY X, it wasn't even him who picked up on the difference in IQ.

Personally i don't think there's anything shady going on here and there's most likely a simple explanation for it.

Spot on. I made this thread to stop the owners thread going off topic and I did say I don't mind the discussion but I did make the video to show performance and nothing more and nothing less.

Anyways, enough of that and time to do some Titan X recording with SP

Mauller · 3 Jul 2015 at 22:17

Just going to post again. But did you refresh your game settings between card swaps greg? was wondering if it could be a game issue in that regards which caused the TX issues.

loftie · 3 Jul 2015 at 22:22

Why is it we can't stay on topic and actually discuss the IQ from both sides hmmm? Why is it instead of discussing what might be causing these differences a lot of people have decided to rubbish Greg's videos?

If Greg has missed something out, not thought to check/do something be helpful and suggest something. Don't just say how rubbish it is because a) nothing is learned b) it's not the case c) you're just being a d***.

If AMD have colour issues, it'd be interesting to know. If Nvidia have LoD issues it'd be interesting to know. And if neither hold true and something wonky is going on with Gregs PC, I'm sure he'd be interested to know!

Thank you Greg for taking the time to do these videos.

provost · 3 Jul 2015 at 22:23

Good work Gregster

But still like your other avatar better.... Lol

humbug · 3 Jul 2015 at 22:23

just watched your review Greg, nicely done

https://www.youtube.com/watch?v=1Nye2rhsA94

Gregster · 3 Jul 2015 at 22:25

Typical!!! want to get on with recording and battlelog is being a ****

Kaapstad · 3 Jul 2015 at 22:51

humbug said:
just watched your review Greg, nicely done

https://www.youtube.com/watch?v=1Nye2rhsA94

That is what it is all about , we all like videogaming and we are all on the same side.

Silent_Scone · 3 Jul 2015 at 22:53

At least Greg has bothered to try and debunk it rather than aimlessly going in circles speculating probable cause. Personally I don't think there's much worth looking in to here, but there aren't many end users with both GPU. On Sunday I will attempt to capture screenshots at 4K in various games in different scenarios and see if there's anything of note with fidelity.

Uncompressed screen grabs should show any differences if there are any. Image fidelity can be directly manipulated through driver presets. For example texture filtering quality if predetermined at it's default maybe very marginally disimilar to the other vendors. LOD parameters may also be swayed in either direction. I think both parties have learnt from age old mistakes in compromising IQ for performance, and it's certainly a lot easier to pick these things up now than it was 5 or 10 years ago. If it wasn't, then things like overwriting tessfactors and geometry which is a direct result of manipulating source code would be done far more often without user intervention. This is applicable to all aspects of the PC industry, people are not as easily lead.

GoogalyMoogaly · 3 Jul 2015 at 22:54

Is there another game that could be used for testing that has a benchmark in it?
Would make it easier to compare and get screenshots in the same place.

Or is it just BF4 that this happens in?
If not it might be interesting to see if/how different games are affected.

You're not busy or anything this weekend are you Greg?

humbug · 3 Jul 2015 at 22:56

Kaapstad said:
That is what it is all about , we all like videogaming and we are all on the same side.

Yes, Hardware enthusiast brethren, we should beat on Console scum <joke, bad tasteless joke

not each-other. Even if 75% of us have chosen the wrong colour GPU, great joke

GoogalyMoogaly · 3 Jul 2015 at 22:59

humbug said:
Yes, Hardware enthusiast brethren, we should beat on Console scum <joke, bad tasteless joke not each-other. Even if 75% of us have chosen the wrong colour GPU, great joke

I didn't realise AMD was that popular!?
Unless you're just talking about this forum, then I'd say you're underestimating!

Kaapstad · 3 Jul 2015 at 23:00

Silent_Scone said:
At least Greg has bothered to try and debunk it rather than aimlessly going in circles speculating probable cause. Personally I don't think there's much worth looking in to here, but there aren't many end users with both GPU. On Sunday I will attempt to capture screenshots at 4K in various games in different scenarios and see if there's anything of note with fidelity.

Uncompressed screen grabs should show any differences if there are any. Image fidelity can be directly manipulated through driver presets. For example texture filtering quality if predetermined at it's default maybe very marginally disimilar to the other vendors. LOD parameters may also be swayed in either direction. I think both parties have learnt from age old mistakes in compromising IQ for performance, and it's certainly a lot easier to pick these things up now than it was 5 or 10 years ago. If it wasn't, then things like overwriting tessfactors and geometry which is a direct result of manipulating source code would be done far more often without user intervention. This is applicable to all aspects of the PC industry, people are not as easily lead.

Good post

Always keep an open mind.

humbug · 3 Jul 2015 at 23:01

GoogalyMoogaly said:
I didn't realise AMD was that popular!?
Unless you're just talking about this forum, then I'd say you're underestimating!

Gregster · 3 Jul 2015 at 23:02

Right, recording is done and now rendering. I am sure someone will pick fault with it but I have had a good day and I have tomorrow off, so go for it

It will take about 20 mins to encode and 10 mins to upload. I use a double pass and a few other settings for my high quality vids and it will be 60 fps, so should be easier to see if there is anything amiss. I forgot to take screenshots (sorry) but can do again if people want.

Gerard · 3 Jul 2015 at 23:06

Gregster · 3 Jul 2015 at 23:09

Things take forever to do when you watch them :mad:

humbug · 3 Jul 2015 at 23:12

Gregster said:
Things take forever to do when you watch them

Take your time. i don't have the time to pick over it yet anyway, i need to get Cryengine working in Linux or Pete will be very cross with me

ashmanuk67 · 3 Jul 2015 at 23:12

Greg could it be this? (Texture Filtering - Negative LOD bias = CLAMP

Default is ALLOW

Gregster · 3 Jul 2015 at 23:15

ashmanuk67 said:
Greg could it be this? (Texture Filtering - Negative LOD bias = CLAMP

Default is ALLOW

I will check in a bit. I never touched the NVCP, as like I said, the TX was only going to be used for 2 hours max to record. The only thing I did with the CCC was alter the colour slightly.