This article make ATI look pretty bad. Unfortunately, it doesn't reconcile with other reviews out there. Given that those reviews did do in depth studies, it looks like the results presented here may involve a selection of non-representative tasks. Sloppy, sloppy work: any good tester knows that you need a large, diverse sample before you draw strong conclusions. Usually Anand does a great job, so I'll let this one slide. But in the future, don't publish half-assed tests like the half-assed results are something we should expect across the board.
That look is from Tom's Hardware Guide, which I have come to trust less exponentially over the years. It looks at quality somewhat but does not place the importance on quality that you have. They are also very capricious about keeping score between ATI and NVidia, giving and taking away points "just because."
Importantly, however, they do use quite sound logic when explaining their choice of clips used.
This makes me think that something is seriously wrong... especially when the times for GPU vs CPU encoding differ so much from yours... a metric that is quality-agnostic.
So what I think people are saying is that we have in front of us several reviews from other sites on one hand, and your review on the other.
However, even being an admitted ATI fanboy who does a TON of video transcoding, I hesitate to purchase Espresso because I give greater weight to Anandtech reviews.
As a result, your review, and the weight I accord it (especially after learning that you used the 9.5 hotfix drivers, which shocked me as I was sure that was the reason for the discrepancy)... in any case it is only this review that is preventing me from purchasing Espresso and if there is any chance that by revisiting the matter you might find cause to justify the 40 USD cost, then we (or at least I) would appreciate it.
Either way, I have appreciated your responsiveness to our queries.
note: by "trust less exponentially" I mean I trust THG less and less as time goes on... but despite capricious and seemingly bias results, the video clips and codecs used in this case seem well thought out (and for the most part available to anyone)
loox,
If you truly encode, etc you would already know that it's impossible to get the same type of artifact in the exact same location using Avivo and Media Espresso. I've used them both since this blog was posted and I have not been able to replicate this at all.
Further more, we are now using Cat 9.6 using avivo 9.6 which IMO makes this blog now moot at best. Check out their release notes they've addressed their Avivo program.
The Expresso may have been using different set of decoder-encoder for these 3 comparisons you made (as what you explained). The CPU encoding in your review may be using non-Avivo encoder. Did you try to transcode using Avivo converter with GPU acceleration turned off (i.e. CPU + Avivo encoder) and compare the quality result with GPU + Avivo encoder? If no artifact, then this would mean the GPU portion of the encoding process is the culprit i.e. Radeon GPU transcoding is BS.
I've been using Espresso Media Show for sometime now and never seen any artifacts like that. Furthermore, I've never seen a program like this that always artifacts in the same spot. So, this isn't a program with others who use the program.
Why would anyone waster time and effort on GPU encoders that are optimized for speed not quality.
If you match the settings those GPU encoders use x264 would be just as fast only x264 was built for quality first and speed second.
I look forward to seeing a proper H.264 encoder like x264 using Direct-X computer shaders or even OpenCL in a meaningful way but for now these applictions are a waste of time and money you get better speed and quality from free x264 encoders like Ripbot264, AutoMKV, XviD4PSP etc.
Today's GPU are not encoders. They are stream processors that are designed to efficiently compute formulas. These stream processors can handle any mathematical problems. They can be multimedia decoders and encoders, predict the reaction of a chemical reaction, artificial intelligence, and other problems.
The visual quality depends on the software since stream processing still has the problem with garbage in, garbage out (GIGO).
There are commercial encoders that uses stream processors. Unfortunately, they cost over US $1000 -- the good ones. There is dirac which is a government run open source project.
ATI hardware is at the mercy of its software. IMHO, The software developers at ATI should be fired. Then higher new fresh software developers that completely writes drivers from the ground and up. The open source community is writing software for ATI cards from the ground and up and it is a lot better.
Are the differences due to ET using only 24p sources and using 60i sources here? Or is it due to Catalyst 9.5 hotfix (which is what the websites that got it to work used)? It would be nice if someone did a meta-analysis of the results to find out what the problem is. A test of the type "we tried it, but couldn't get it to work" isn't very useful. Keeping it going until you get to the bottom of the problem is a lot more useful. (Even if the actual problem isn't solved.)
We did use the 9.5 hotfix, and yet these are the results we got.
The interlaces source I used was the MacGyver episode. I had problems with multiple sources including progressive 60fps sources (the fraps video). It's not a "we tried it and couldn't get it to work" -- this is a "this implementation does not work" situation.
The bottom of the problem is that AMD has not implemented a quality transcoding algorithm either for their own AVIVO converter or for 3rd party applications.
They don't look at the image quality but concentrate on performance on different platforms. Really interesting results.
About the image quality:
Well I remember reading that Cyberlink and ArcSoft are just using the ATI Stream Transcoding runtime. That's why both AVIVO Video Converter and Espresso exhibit the same image quality issues. This runtime is built by ATI, so all programs using this runtime will have the same image quality.
To what extend ATI (through drivers) or 3rd party companies (through updates) will update this runtime, will be interesting.
I believe that ATI hasn't been all bad here. Most of the people banish their old cards, but you have to remember the days of the RAGE chips, when they introduced a whole VIDEO chip on the cards, just to accelerate TV recording. Back in the days MPEG-2 was a whole new concept to Windows PCs and they were there. Now, clearly, nVidia is a bit faster at DX10 generation chips, but both are not really a revolution. ATI could patch things up, lose another 5-10% to nVidia and the chip would still be a good investment for the mass user with a 2-4 Core CPU, which I don't expect to use extensive decoding with this tools yet, as multimedia is not generally 1080p. DX11, Windows 7 - then yes we are talking. New performance of the current crop CPU's, NEW GPU's ... this is yet to be seen.
Saying that - nVidia is still a good option, but their chip is a bit overpriced as well. An nVidia user.
I was just playing with this stuff the other night. Espresso does a nice job but can't get it to work right on my ATI hardware. My 4830 is in a box running Win7 x64 hardware acceleration isn't even available despite the new hotfix drivers etc... My old box has XP Pro and has a 3870 no dice on that one for hardware acceleration either. Seems the hardware isn't supported. Maybe someday... Ne way my rinky dink 8600M GT in my Laptop running Win7 x32 works perfect. The quality is terrific and the encode times are cut in half. after this tidbit I'm starting to wonder if there is an advantage here using a high-end card with this software. An update run with some more conservative hardware would be great. Perhaps a 8800GT or 9600 GT, maybe a E8500 or E6600 CPUs as well... Funny how the ATI cards tear through F@H packets but don't fare as well in Espresso...
"We plan on taking a more indepth look at GPU accelerated transcoding when we can make more objective comparisons and draw more useful conclusions."
Translation: Nvidia wins and ati sucks so badly, we can't really cheer for the red cards or even make "a comparison" declaring ati the winner, as we so often try to do. We'd rather not declare Nvidia the absolute king in this area - and really, who cares - physx, cuda, forced sli, built in game presets, ambient occlusion - it all sucks because nvidia raped us, and so does this because - well we want it so badly, but ati our red wondercard company blew it, so we'll pretend we could care less.
-
"Right now everything is just too up in the air."
Translation: ati really sucks at this - why is this happening to us !!?? please forgive us honorable and worshipped red masters!
-
"We're hoping that the emergence of OpenCL will help to unify the implementations of developers to the point where we can have something"
Translation: Dear GOD help us get ati on it's feet- something, ANYTHING to force Nvidia down into the gutter and pull ati up ! Please, please !! Hope for change !
-
"that is equivalent (or at least much closer)"
Translation: It's so bad we don't think an ati card can do it, they just don't have the "brute force" we dissed nvidia so much for, but we can't bring ourselves to SAY IT, because we love the red soooooo much. We'll just shut up and HOPE HOPE HOPE !
-
"in output quality in order to better evaluate the capabilities of the hardware when transcoding video."
Translation: The capabilities of NVidia are kicking red A-ti, but with the correct miracle, we'll be able to claim ati can do this or that "better", then we can spin it into buy a red card because you can't live without this !
Until then, who gives a fluke about moving video about or improving it ? Noone, you don't need, we don't need it, noone uses it, and you should NEVER base your purchase decision on nvidia's selfish implementation because they're just mean and greedy.
" AHHH-MMEEENNN "
--------------------
Thanks so much - another reason to throw red cards in the trash where they belong - in front of all the fans who have that hate filled chip in their soul against nvidia.(and boy do they ever come out of the closet and tel me why they do)
TOO BAD NVIDIA WINS AGAIN, WINS BIG, WINS ABSOLUTELY, AND CRUSHES THE little red loser !
----------------
BUY NVIDIA if you have a lick of sense.
we did try two different 4890 cards with the same result. it really doesn't look like a temporal anomaly caused by some hardware defect as the results seem too consistent (same problems every time).
I think that the benchmarks clearly show that GPU based calculations are barely giving a factor of two speedup, never mind an order of magnitude.
For cards with over 1TFOPS of computational power on board (10-50x more than a CPU) they sure do suck. It goes to show that having that power available means nothing if you can't get the data there effectively or something else is acting as a roadblock.
AMD need to get some video experts in to sort their transcoding out. Right now it's an embarrassment to have such poor performance and quality. The PR looks good - use the on-board video decoder to decode the MPEG2/H264/VC1/etc, and then pass the decoded video data on within the GPU to the stream-based encoder. CPU involvement should be minimal. Maybe they need more memory cache on the GPUs for it to work more effectively...
Can someone please describe a situation, when it matters if you encode a video 2 minutes faster (at home)?
:D
I'm not against technology advancements, but it shouldn't be a reason of purchase for the average folks and even very important.
On the other hand, obviously the quality should be brilliant...I can't understand this, a few years ago ATI was the leader in this. maybe they have fired some good slaves by mistake :)
I don't use my Archos 5 for video, but lots of users do, and a 250GB HDD can handle a lot of optimized videos. I'm sure some would be willing to spend some money on an application if it reduced the transcoding time for their collection from weeks to days.
I've used Media Show Espresso for sometime now with my 4850 and never come across IQ issues like this before. Honestly, I cannot relate to this brief article.
I tried getting this to work for two days on my 4850. Then the reviews for it came out and I realized it was useless. This was a major influence on me switching to Nvidia this time. I have a real problem with people wasting my time and basically lying to me.
It was rubbish back in the X1xxx/R520 days when they first introduced it, and it's still rubbish now. It looks as if they haven't done any work on it whatsoever.
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
32 Comments
Back to Article
mospeada - Thursday, June 18, 2009 - link
GPU transcoding not up to the par, yet. See http://www.pcgeek.my">http://www.pcgeek.myphilosofool - Monday, June 15, 2009 - link
This article make ATI look pretty bad. Unfortunately, it doesn't reconcile with other reviews out there. Given that those reviews did do in depth studies, it looks like the results presented here may involve a selection of non-representative tasks. Sloppy, sloppy work: any good tester knows that you need a large, diverse sample before you draw strong conclusions. Usually Anand does a great job, so I'll let this one slide. But in the future, don't publish half-assed tests like the half-assed results are something we should expect across the board.DerekWilson - Tuesday, June 16, 2009 - link
could you point me to those in-depth studies that look at image quality? i'm interested in what clips they used for comparison.loox - Tuesday, June 16, 2009 - link
http://www.tomshardware.com/reviews/amd-stream-gpg...">http://www.tomshardware.com/reviews/amd-stream-gpg...That look is from Tom's Hardware Guide, which I have come to trust less exponentially over the years. It looks at quality somewhat but does not place the importance on quality that you have. They are also very capricious about keeping score between ATI and NVidia, giving and taking away points "just because."
Importantly, however, they do use quite sound logic when explaining their choice of clips used.
This makes me think that something is seriously wrong... especially when the times for GPU vs CPU encoding differ so much from yours... a metric that is quality-agnostic.
So what I think people are saying is that we have in front of us several reviews from other sites on one hand, and your review on the other.
However, even being an admitted ATI fanboy who does a TON of video transcoding, I hesitate to purchase Espresso because I give greater weight to Anandtech reviews.
As a result, your review, and the weight I accord it (especially after learning that you used the 9.5 hotfix drivers, which shocked me as I was sure that was the reason for the discrepancy)... in any case it is only this review that is preventing me from purchasing Espresso and if there is any chance that by revisiting the matter you might find cause to justify the 40 USD cost, then we (or at least I) would appreciate it.
Either way, I have appreciated your responsiveness to our queries.
loox - Tuesday, June 16, 2009 - link
note: by "trust less exponentially" I mean I trust THG less and less as time goes on... but despite capricious and seemingly bias results, the video clips and codecs used in this case seem well thought out (and for the most part available to anyone)CPUGuy - Thursday, June 18, 2009 - link
loox,If you truly encode, etc you would already know that it's impossible to get the same type of artifact in the exact same location using Avivo and Media Espresso. I've used them both since this blog was posted and I have not been able to replicate this at all.
Further more, we are now using Cat 9.6 using avivo 9.6 which IMO makes this blog now moot at best. Check out their release notes they've addressed their Avivo program.
mospeada - Sunday, June 14, 2009 - link
The Expresso may have been using different set of decoder-encoder for these 3 comparisons you made (as what you explained). The CPU encoding in your review may be using non-Avivo encoder. Did you try to transcode using Avivo converter with GPU acceleration turned off (i.e. CPU + Avivo encoder) and compare the quality result with GPU + Avivo encoder? If no artifact, then this would mean the GPU portion of the encoding process is the culprit i.e. Radeon GPU transcoding is BS.CPUGuy - Saturday, June 13, 2009 - link
I've been using Espresso Media Show for sometime now and never seen any artifacts like that. Furthermore, I've never seen a program like this that always artifacts in the same spot. So, this isn't a program with others who use the program.Rainman200 - Friday, June 12, 2009 - link
Why would anyone waster time and effort on GPU encoders that are optimized for speed not quality.If you match the settings those GPU encoders use x264 would be just as fast only x264 was built for quality first and speed second.
I look forward to seeing a proper H.264 encoder like x264 using Direct-X computer shaders or even OpenCL in a meaningful way but for now these applictions are a waste of time and money you get better speed and quality from free x264 encoders like Ripbot264, AutoMKV, XviD4PSP etc.
jmurbank - Sunday, June 14, 2009 - link
Today's GPU are not encoders. They are stream processors that are designed to efficiently compute formulas. These stream processors can handle any mathematical problems. They can be multimedia decoders and encoders, predict the reaction of a chemical reaction, artificial intelligence, and other problems.The visual quality depends on the software since stream processing still has the problem with garbage in, garbage out (GIGO).
There are commercial encoders that uses stream processors. Unfortunately, they cost over US $1000 -- the good ones. There is dirac which is a government run open source project.
ATI hardware is at the mercy of its software. IMHO, The software developers at ATI should be fired. Then higher new fresh software developers that completely writes drivers from the ground and up. The open source community is writing software for ATI cards from the ground and up and it is a lot better.
randomname - Friday, June 12, 2009 - link
It is sometimes very difficult to reconcile the opinions of different websites.http://www.extremetech.com/article2/0,2845,2348555...">http://www.extremetech.com/article2/0,2845,2348555...
First page:
http://www.extremetech.com/article2/0,2845,2348550...">http://www.extremetech.com/article2/0,2845,2348550...
Are the differences due to ET using only 24p sources and using 60i sources here? Or is it due to Catalyst 9.5 hotfix (which is what the websites that got it to work used)? It would be nice if someone did a meta-analysis of the results to find out what the problem is. A test of the type "we tried it, but couldn't get it to work" isn't very useful. Keeping it going until you get to the bottom of the problem is a lot more useful. (Even if the actual problem isn't solved.)
DerekWilson - Saturday, June 13, 2009 - link
We did use the 9.5 hotfix, and yet these are the results we got.The interlaces source I used was the MacGyver episode. I had problems with multiple sources including progressive 60fps sources (the fraps video). It's not a "we tried it and couldn't get it to work" -- this is a "this implementation does not work" situation.
The bottom of the problem is that AMD has not implemented a quality transcoding algorithm either for their own AVIVO converter or for 3rd party applications.
The problem will be solved when AMD solves it.
MrMilli - Thursday, June 11, 2009 - link
This review is really interesting:http://legitreviews.com/article/978/1/">http://legitreviews.com/article/978/1/
They don't look at the image quality but concentrate on performance on different platforms. Really interesting results.
About the image quality:
Well I remember reading that Cyberlink and ArcSoft are just using the ATI Stream Transcoding runtime. That's why both AVIVO Video Converter and Espresso exhibit the same image quality issues. This runtime is built by ATI, so all programs using this runtime will have the same image quality.
To what extend ATI (through drivers) or 3rd party companies (through updates) will update this runtime, will be interesting.
koss - Thursday, June 11, 2009 - link
I believe that ATI hasn't been all bad here. Most of the people banish their old cards, but you have to remember the days of the RAGE chips, when they introduced a whole VIDEO chip on the cards, just to accelerate TV recording. Back in the days MPEG-2 was a whole new concept to Windows PCs and they were there. Now, clearly, nVidia is a bit faster at DX10 generation chips, but both are not really a revolution. ATI could patch things up, lose another 5-10% to nVidia and the chip would still be a good investment for the mass user with a 2-4 Core CPU, which I don't expect to use extensive decoding with this tools yet, as multimedia is not generally 1080p. DX11, Windows 7 - then yes we are talking. New performance of the current crop CPU's, NEW GPU's ... this is yet to be seen.Saying that - nVidia is still a good option, but their chip is a bit overpriced as well. An nVidia user.
teohhanhui - Thursday, June 11, 2009 - link
Maybe they can just contribute to the x264 project to make it happen. That would be really nice.Einy0 - Thursday, June 11, 2009 - link
I was just playing with this stuff the other night. Espresso does a nice job but can't get it to work right on my ATI hardware. My 4830 is in a box running Win7 x64 hardware acceleration isn't even available despite the new hotfix drivers etc... My old box has XP Pro and has a 3870 no dice on that one for hardware acceleration either. Seems the hardware isn't supported. Maybe someday... Ne way my rinky dink 8600M GT in my Laptop running Win7 x32 works perfect. The quality is terrific and the encode times are cut in half. after this tidbit I'm starting to wonder if there is an advantage here using a high-end card with this software. An update run with some more conservative hardware would be great. Perhaps a 8800GT or 9600 GT, maybe a E8500 or E6600 CPUs as well... Funny how the ATI cards tear through F@H packets but don't fare as well in Espresso...hybrid2d4x4 - Thursday, June 11, 2009 - link
Thanks for the update, but did you guys use more than one vid card in your testing? Could the artifacting be a hardware issue with the card you have?kn00tcn - Saturday, June 20, 2009 - link
question is what is the player/decoder?i have exported h.264 out of aftereffects/premiere, & ffdshow gives blocky artifacts, while coreavc & youtube do not (& i dont touch vlc)
(& worse, AE or premiere cant even import what they exported...)
i have a workflow with MEGui for encoding my fraps videos, so untill that (or x264 in general) can utilize the gpu, i'm stuck with cpu encodes
SiliconDoc - Sunday, July 5, 2009 - link
"We plan on taking a more indepth look at GPU accelerated transcoding when we can make more objective comparisons and draw more useful conclusions."Translation: Nvidia wins and ati sucks so badly, we can't really cheer for the red cards or even make "a comparison" declaring ati the winner, as we so often try to do. We'd rather not declare Nvidia the absolute king in this area - and really, who cares - physx, cuda, forced sli, built in game presets, ambient occlusion - it all sucks because nvidia raped us, and so does this because - well we want it so badly, but ati our red wondercard company blew it, so we'll pretend we could care less.
-
"Right now everything is just too up in the air."
Translation: ati really sucks at this - why is this happening to us !!?? please forgive us honorable and worshipped red masters!
-
"We're hoping that the emergence of OpenCL will help to unify the implementations of developers to the point where we can have something"
Translation: Dear GOD help us get ati on it's feet- something, ANYTHING to force Nvidia down into the gutter and pull ati up ! Please, please !! Hope for change !
-
"that is equivalent (or at least much closer)"
Translation: It's so bad we don't think an ati card can do it, they just don't have the "brute force" we dissed nvidia so much for, but we can't bring ourselves to SAY IT, because we love the red soooooo much. We'll just shut up and HOPE HOPE HOPE !
-
"in output quality in order to better evaluate the capabilities of the hardware when transcoding video."
Translation: The capabilities of NVidia are kicking red A-ti, but with the correct miracle, we'll be able to claim ati can do this or that "better", then we can spin it into buy a red card because you can't live without this !
Until then, who gives a fluke about moving video about or improving it ? Noone, you don't need, we don't need it, noone uses it, and you should NEVER base your purchase decision on nvidia's selfish implementation because they're just mean and greedy.
" AHHH-MMEEENNN "
--------------------
Thanks so much - another reason to throw red cards in the trash where they belong - in front of all the fans who have that hate filled chip in their soul against nvidia.(and boy do they ever come out of the closet and tel me why they do)
TOO BAD NVIDIA WINS AGAIN, WINS BIG, WINS ABSOLUTELY, AND CRUSHES THE little red loser !
----------------
BUY NVIDIA if you have a lick of sense.
DerekWilson - Thursday, June 11, 2009 - link
we did try two different 4890 cards with the same result. it really doesn't look like a temporal anomaly caused by some hardware defect as the results seem too consistent (same problems every time).psychobriggsy - Thursday, June 11, 2009 - link
I think that the benchmarks clearly show that GPU based calculations are barely giving a factor of two speedup, never mind an order of magnitude.For cards with over 1TFOPS of computational power on board (10-50x more than a CPU) they sure do suck. It goes to show that having that power available means nothing if you can't get the data there effectively or something else is acting as a roadblock.
AMD need to get some video experts in to sort their transcoding out. Right now it's an embarrassment to have such poor performance and quality. The PR looks good - use the on-board video decoder to decode the MPEG2/H264/VC1/etc, and then pass the decoded video data on within the GPU to the stream-based encoder. CPU involvement should be minimal. Maybe they need more memory cache on the GPUs for it to work more effectively...
lez - Thursday, June 11, 2009 - link
Can someone please describe a situation, when it matters if you encode a video 2 minutes faster (at home)?:D
I'm not against technology advancements, but it shouldn't be a reason of purchase for the average folks and even very important.
On the other hand, obviously the quality should be brilliant...I can't understand this, a few years ago ATI was the leader in this. maybe they have fired some good slaves by mistake :)
strikeback03 - Friday, June 12, 2009 - link
I don't use my Archos 5 for video, but lots of users do, and a 250GB HDD can handle a lot of optimized videos. I'm sure some would be willing to spend some money on an application if it reduced the transcoding time for their collection from weeks to days.Aquila76 - Thursday, June 11, 2009 - link
Think of it as encoding in half the time, and then extrapolate that to a full length movie.CPUGuy - Thursday, June 11, 2009 - link
I've used Media Show Espresso for sometime now with my 4850 and never come across IQ issues like this before. Honestly, I cannot relate to this brief article.aeternitas - Thursday, June 11, 2009 - link
Its trash.dryloch - Thursday, June 11, 2009 - link
I tried getting this to work for two days on my 4850. Then the reviews for it came out and I realized it was useless. This was a major influence on me switching to Nvidia this time. I have a real problem with people wasting my time and basically lying to me.KayDat - Thursday, June 11, 2009 - link
It was rubbish back in the X1xxx/R520 days when they first introduced it, and it's still rubbish now. It looks as if they haven't done any work on it whatsoever.becklewis - Thursday, June 24, 2010 - link
well done. i have a bunch of video needed to be converted for my new bought ipod. thanks for sharing,i will give it a try.here is the one i uesd
http://www.best-video-converter.net
becklewis - Thursday, June 24, 2010 - link
why cant edit my comment?<a hreff="http://www.best-video-converter.net "> AVS video converter </a> is a good one.
becklewis - Thursday, June 24, 2010 - link
why cant edit my comment?http://www.best-video-converter.net"> AVS video converter is a good one.
karoldude - Monday, June 28, 2010 - link
cool staff ,thanks for sharing . here is a video converter review , i dont know how to choose , need some help.
http://www.best-video-converter.net