[ventro]

I see that the new 5.13 drivers are out.

https://support.ati.com/ics/support/...questionID=640


I was wondering, does all that extra video stuff like deinterlacing and image sharpening work on a 9800Pro or is it only for the X1000series? And if that is the case, is there a hack or something to get it working on a 9800 Pro?

 

[miketech]

All those new features apply to the X1xxxx cards only. The older cards wil do adaptive deinterlacing w/ Cat 5.9 or above.

 

[stanger89]

Quote:

HQV Video Benchmark Support The Catalyst® software suite 5.13 introduces full support for the HQV DVD Benchmark. These driver tweaks provide a better DVD image quality, and higher benchmark scores.

ROFL

 

[Tulli]

It's in your signature, Stanger89 ... Ati beat you!!!

 

[DustinF]

Quote:

Originally Posted by stanger89 ROFL

Firing Squad review of 5.13


.

 

[stanger89]

ROFL is because they state they optimized their drivers for that benchmark. If they did that with 3Dmark, they'd be flamed to ashes on the gaming sites.


The only other comments I'll make are:

Something's not right with one of the configs, the levels between the ATI and nVidia screenshots are not the same, one of them must be configured wrong.


They're comparing the flagship X1800 vs the low-midrange 6600. Don't know if the 7800 would do better, but it would remove doubt.

 

[Zissou]

Quote:

Originally Posted by stanger89 ROFL is because they state they optimized their drivers for that benchmark. If they did that with 3Dmark, they'd be flamed to ashes on the gaming sites. The only other comments I'll make are: Something's not right with one of the configs, the levels between the ATI and nVidia screenshots are not the same, one of them must be configured wrong. They're comparing the flagship X1800 vs the low-midrange 6600. Don't know if the 7800 would do better, but it would remove doubt.

Full support not optimizations. Comparing 3dmark optimizing to to proper and new deinterlacing strategies is ridiculous. You obviously have great concern for HQV as you refer to it in your sig. Why don't you buy a 7800 and 1800 and compare them as I have? Until then you're merely spreading FUD and innuendo.

 

[Tulli]

Stanger89:


But then the HQV Benchmark sucks in itself. For a good benchmark should be a good objective measure of what specifically tests, and could certainly be used as a tool or directive for testing products. I mean for instance, your graphic engine shouldn't give you bad 3Dmark. I can't imagine ATI nor NVidia not consideing that benchmark for their cards/drivers.


As for the X1800/6000GT issue in the Firingsquads review, it's been already answered by the writer of that review in this thread:

http://www.avsforum.com/avs-vb/showthread.php?t=617940


I quote:

Quote:

PureVideo is a two-part solution. There's a dedicated programmable VPU, and the 3D graphics part. The algorithms originally required the processing power of the 7800GTX's VPU which is more powerful than the 6600GT's. However, NVIDIA determined that they could use the same algorithms by using the 3D graphics portion of the 6600GT to "make up the difference.

Bottom line for me anyway is that deinterlacing is such a tricky business that it can't really be solved by just one solution. So the best I expect is some decent alternative. If ATI is able to add some "good" film-detection - not that hard indeed - that's fine to me. I don't watch that much video stuff anyway.

 

[stanger89]

Quote:

Originally Posted by Zissou Full support not optimizations. Comparing 3dmark optimizing to to proper and new deinterlacing strategies is ridiculous.

Not really, I'm just pointing out that they stated they were optimizing for a benchmark. Seems funny to me. If they'd have left the HQV stuff out of and just said "These driver tweaks improve DVD quality" there'd have been no comment from me. It was the specifically calling out the HQV benchmark that struck me as funny.

Quote:

You obviously have great concern for HQV as you refer to it in your sig.

I link it because (flawed as it is) it's the closest thing to an objective benchmark for video quality we've got.

Quote:

Why don't you buy a 7800 and 1800 and compare them as I have?

Mainly because I don't care to, I'll get a 7800 as soon as I can find one passively cooled.

Quote:

Until then you're merely spreading FUD and innuendo.

Look I don't have a problem with ATI doing better than nVidia, it's about time, and I'm surprised it took them so long, given their past superiority. I just pointed out it was funny that they stated they optimized for a benchmark.

 

[Zissou]

Quote:

Originally Posted by stanger89 Not really, I'm just pointing out that they stated they were optimizing for a benchmark. Seems funny to me. If they'd have left the HQV stuff out of and just said "These driver tweaks improve DVD quality" there'd have been no comment from me. It was the specifically calling out the HQV benchmark that struck me as funny. I link it because (flawed as it is) it's the closest thing to an objective benchmark for video quality we've got. Mainly because I don't care to, I'll get a 7800 as soon as I can find one passively cooled. Look I don't have a problem with ATI doing better than nVidia, it's about time, and I'm surprised it took them so long, given their past superiority. I just pointed out it was funny that they stated they optimized for a benchmark.

Again they did not state they were optimizing for it, those are your words and quite incorrect and misleading. You can state over and over again they optimized for it when in fact they didn't they're simply using vastly improved deinterlacing algorithms. You're right it's the closest thing to an objective benchmark and ATI leaves nVidia in the dust now. End of story.

 

[stanger89]

Quote:

Originally Posted by Zissou Again they did not state they were optimizing for it, those are your words and quite incorrect and misleading. You can state over and over again they optimized for it when in fact they didn't they're simply using vastly improved deinterlacing algorithms.

I am not saying they did or did not optomize for it, I'm say they said they optimized for it. What I'm saying is that there are much better ways they could have stated the same thing. eg:

Improved DVD Image Quality


The Catalyst® software suite 5.13 introduces driver tweaks provide a better DVD image quality. As a result, there is now full support for the HQV DVD Benchmark and higher benchmark scores.


It's like for gaming, saying something like this:


Improved gaming performance


The blah-blah-blah introduces driver tweaks that provide improved performance for D3D games. As a result 3DMark is now fully supported and scores are improved.


vs if they were to say:


3DMark 2006 Benchmark support


The blah-blah-blah introduces full support for the 3DMark 2006 Benchmark. These driver tweaks provide a higher D3D performance, and higher benchmark scores.


ATI (or nVidia) would be flamed if they were to put something like that in their drivers/release notes for a gaming benchmark. They should be smarter than that.


It's the focus on improved HQV performance that is what I was commenting on.

Quote:

You're right it's the closest thing to an objective benchmark and ATI leaves nVidia in the dust now. End of story.

I never contested that.

 

[Naylia]

I agree, the second sentece shouldn't have been included in their driver release notes...just leave it at

Quote:

The Catalyst® software suite 5.13 introduces driver tweaks provide a better DVD image quality.

and stop there. Then allow the benchmarks themselves to review the software and do their job of announcing improved scores.



On a good note...it sounds like the improved DVD image quality is applicable to all the ATI cards that fall under the drivers since there are no particular models focused on in the section. Is anyone able to see a visable difference?

 

[trvtec]

I installed the beta 5.13's and I was not happy with the results in ZP/FFDShow/LimitedSharpen. Since they still haven't fixed the DVI black-out issue, I rolled back to 5.6 and am much happier with the picture. I didn't play with it much once I discovered that the DVI issue still exists, so that could explain my findings. I guess you could say I didn't give it much of a chance since I prefer not unplugging my TV all the time just to get a picture from the PC.


Todd.

 

[DustinF]

Quote:

Originally Posted by trvtec I installed the beta 5.13's and I was not happy with the results in ZP/FFDShow/LimitedSharpen. Since they still haven't fixed the DVI black-out issue, I rolled back to 5.6 and am much happier with the picture. I didn't play with it much once I discovered that the DVI issue still exists, so that could explain my findings. I guess you could say I didn't give it much of a chance since I prefer not unplugging my TV all the time just to get a picture from the PC. Todd.

You need an X1*** card to see any benefit from the new driver.


I have an x800xl and it did nothing.

 

[miketech]

I installed 5.13 last night on my room PC. 9800pro/Olevia 26"/ 1280x768/XP2800+/Dscaler5007/ZP 4.51.

I don't have the HQV disk but I usually test w/ one of my worst flagged DVD.(Dave Mason Live). I selected vector adaptive deinterlacing(w/c never worked before on pre 5.13) and it actually worked this time. And I did see PQ improvement to my surprise. I know this is very subjective and I only tested for 30 min. I will try to do the same on my main system this weekend(X800XL).

If others need to try it, do not use ffdshow and use the Dscaler codec set to NV12. No deinterlacer w/ Dscaler so you are sure this is hardware at work....

 

[GX-Alan]

Quote:

Something's not right with one of the configs, the levels between the ATI and nVidia screenshots are not the same, one of them must be configured wrong.

That's why you need to read the article. We use real analog VGA capture for our screenshots. These differences in levels are idiosyncracies with the capture hardware.

 

[stanger89]

Sorry, read through the first article, but just flew threw this latest one.

 

[Zissou]

Quote:

Originally Posted by stanger89 Sorry, read through the first article, but just flew threw this latest one.

Which is why you're were spreading mistruths. Next time read before posting.

 

[kschmit2]

Zissou, you sound ridicilous in each and every of your posts.


You must be the smartest guy to have ever hit this platform

 

[Zissou]

Quote:

Originally Posted by kschmit2 Zissou, you sound ridicilous in each and every of your posts. You must be the smartest guy to have ever hit this platform

If you plan on continuing to be a cheerleader for nVidia at least learn to spell ridiculous.