Originally posted by: ronnn
This many threads on the same subject is well known as fodder for flame wars. I stay as unclear as ever.
Originally posted by: cmdrdredd
It's generally better to release a supported driver more often.
I'm not a "he", thank you very much .Originally posted by: nRollo
Originally posted by: ronnn
This many threads on the same subject is well known as fodder for flame wars. I stay as unclear as ever.
I think he's trying to tell you "Argue all you like about drivers and GTX260 vs HD4870 as both are mentioned in the OP, but don't turn this into the 1,245th chapter of "Why I don't like the focus group".
Originally posted by: evolucion8
The Far Cry 2 Anandtech's review shows that the HD 4870 1GB performs dangerously close and often outperforms the GTX 280 which is more powerful than the GTX 260. Not because of this, but in general, the HD 4870 1GB is overall better than the GTX 260 which doen't have DX10.1, VC-1 acceleration, wose 2D image quality, less VRAM, same old technology as the almost 3 years old 8800GTX, the GTX 260 Core 216 is a nice card, and the manufacturer's warranty is nice, but meh. ATi did good this time after two flops.
Originally posted by: AmberClad
I'm not a "he", thank you very much .Originally posted by: nRollo
Originally posted by: ronnn
This many threads on the same subject is well known as fodder for flame wars. I stay as unclear as ever.
I think he's trying to tell you "Argue all you like about drivers and GTX260 vs HD4870 as both are mentioned in the OP, but don't turn this into the 1,245th chapter of "Why I don't like the focus group".
Originally posted by: evolucion8
The Far Cry 2 Anandtech's review shows that the HD 4870 1GB performs dangerously close and often outperforms the GTX 280 which is more powerful than the GTX 260. Not because of this, but in general, the HD 4870 1GB is overall better than the GTX 260 which doen't have DX10.1, VC-1 acceleration, wose 2D image quality, less VRAM, same old technology as the almost 3 years old 8800GTX, the GTX 260 Core 216 is a nice card, and the manufacturer's warranty is nice, but meh. ATi did good this time after two flops.
Originally posted by: nRollo
A woman on the video forum?!
I thought the ladies were way too smart to be around this sort of "Ugh, my video card can beat up your video card" shenanigans and just rolled your eyes at such chest thumping.
Originally posted by: solofly
Originally posted by: evolucion8
The Far Cry 2 Anandtech's review shows that the HD 4870 1GB performs dangerously close and often outperforms the GTX 280 which is more powerful than the GTX 260. Not because of this, but in general, the HD 4870 1GB is overall better than the GTX 260 which doen't have DX10.1, VC-1 acceleration, wose 2D image quality, less VRAM, same old technology as the almost 3 years old 8800GTX, the GTX 260 Core 216 is a nice card, and the manufacturer's warranty is nice, but meh. ATi did good this time after two flops.
You got that right... :thumbsup:
We managed to find out that some of these optimizations are also going to wait for the next Catalyst 9.1 which should probably come sometime next year. The optimizations are going to influence both single board and CrossFire configurations and the only issue is the titles for which Nvidia apparently had more time to optimize than ATI.
Originally posted by: BikeDude
Originally posted by: cmdrdredd
It's generally better to release a supported driver more often.
Although I have never come across software that does not contain bugs, I do feel the idea of monthly bug fixes to be a bit much. That would seem to indicate that something is fundamentally flawed.
OK, I can accept that CF is a complicated piece of technology and that individual fittings is required. It is far from ideal, but probably brought on by circumstances outside AMD's control. But you guys are not discussing that. You are debating bug fixes.
A game like Far Cry 2 is not released without a certain amount of testing. I would assume they did quite a lot of testing to make sure their game ran sufficiently well on the AMD platform. Or did they not? (it would have been much better had Derek Wilson tried to answer that question before pointing the big finger of blame, but at the same time... AMD knows the answer to that question by now and are free to tell us in that case)
I have seen comments by game developers over the years pointing at problems with both manufacturers, so historically I do not think it is not a question of nVidia being less bugprone than AMD.
From my personal experience as an end-user, I know first hand that nVidia's drivers for the past two years have been less than problem-free. BSODs. Freezing problems. Video playback issues. All disappeared eventually after several rounds of bug fixing. Sometimes a driver would introduce a new problem, but eventually it seems they mostly went away. I even have the correct colour profile loaded on my Vista system now. I cannot remember problems with any games at the moment, but I do not get to play many games these days. (and quite frankly, system crashes are more serious than stuff popping or not popping up in a game)
When the 4870 first appeared, I was one of the first who cried "yes please! I want to escape the horrible nVidia drivers!". The discussions sparked by Derek's observations have given me reason for pause.
What I really miss is the opportunity to report bugs. I once tried to tell nVidia that their support for PAE was broken (79.11 is probably still the last version supporting PAE properly), but found that to be impossible. Gainward refused to pass on my bug report and there was no way I could get in touch with nVidia myself.
Personally, I do not want any driver updates at all. I just want everything to work the way it is supposed to. If there are monthly updates to fix bugs over a two year period, without any significant new features being added, then I think an alarm bell should sound somewhere. Or, when nVidia releases a driver that cannot cope with 30 minutes worth of DVD playback (don't worry now, that was a year ago), someone ought to be flogged. Especially when users felt the need to upgrade to that version to fix other issues... That's just messy.
As a software developer, I too am responsible for annoying people with bugs. BUT! My bugs do not cause people to question their system and tricking them into reinstalling everything (somehow many people are still convinced reinstalling the OS will help, because their experience with Win9x is carried over to 2000/XP/Vista). My bugs don't trigger BSODs. Writing kernel mode drivers is a different game altogether. (or should be!)
Originally posted by: nRollo
Originally posted by: Creig
Originally posted by: Kadence
The GTX260 actually seems like a significantly better value to me, and I've also heard lots of stuff recently about Nvidia driver superiority over ATI. However I already have an HD4850, so it doesn't make sense to go Nvidia since I can crossfire my new card with the 4850.
So - is the HD4870 1GB worth it? My 24" monitor's native resolution is 1920x1200. The main game I might be playing is Oblivion, and possibly Diablo 3 in the future.
If both ATI and Nvidia release a driver at the same time and both contain a bug, ATI could have a fix released in as little as 30 days due to their monthly driver releases. With Nvidia, you could easily be waiting 4 to 6 months for the same fix unless they happen to decide to release a "beta" driver. But this is totally at their whim. If it's important to you but not important to them, then you'll probably be waiting awhile. If you do a search you'll find that most other gamers here prefer ATI's monthly release schedule to Nvidia's 4-6 months.
Your logic is flawed in that ATi doesn't fix all bugs in a month, and uses Hotfixes in the same fashion as NVIDIA uses beta releases to get fixes for new games out the door fast when needed.
A great example of this would be Far Cry 2, that over a month after it's release still has horrible driver issues on ATi cards, and never had issues on NVIDIA cards.
You're also ignoring the fact that AnandTech's graphics editor has told us twice now that he considers the monthly WHQL a marketing gimmick and has flat out stated it should be abolished as he's finding more issues with ATis drivers than NVIDIA's.
So while Joe Forum member might be saying "Hot damn! I'm getting monthly drivers- that has to be better!" the guy who runs the forum Joe is on is telling us the monthly driver releases are likely the cause of the problem!
Originally posted by: apoppin
WHO?
Derek Wilson likes the monthly releases; so do most reviewers i have asked. It is far better to know that a driver is to be expected every month and it is easy to set up your reviews based around a schedule. Nvidia LOOKS sloppy in comparison to AMD - it is an issue of perception; and i think Nvidia falls short with their "whenever we feel like it" betas. However, their CEO apparently likes it that way; Nvidia feels that AMD "wastes" nearly a month getting it to MS and back for certification, while Nvidia just releases it as a beta .. so it is that way. Frankly, ,i'd like AMD to do more Betas [like Nvidia] and Nvidia to commit themselves to regular drivers [like AMD] - but i'd like world peace also.
i think Nvidia is recently more out of touch with gamers as they ais for the pro market and wider GPU computational acceptance in the mass market. So i expect it to get not better for us; i hope i am wrong.
So the trade off for going forward with best-case scenario numbers is this page explaining the problems and a plea to AMD to change their approach to driver development for the good of the consumer.
Maintaining a monthly driver release schedule is detrimental to AMD's ability to release quality drivers. This is not the first or only issue we've seen that could have been solved (or at least noticed) by expanded testing that isn't possible with such tight release deadlines.
Maintaining a monthly driver release schedule is detrimental to AMD's ability to release quality drivers. This is not the first or only issue we've seen that could have been solved (or at least noticed) by expanded testing that isn't possible with such tight release deadlines.
The schedule of the drivers is not the issue; it's whether or not they work that counts.Originally posted by: Wreckage
Maintaining a monthly driver release schedule is detrimental to AMD's ability to release quality drivers. This is not the first or only issue we've seen that could have been solved (or at least noticed) by expanded testing that isn't possible with such tight release deadlines.
Wow Apop just got served.
I would say that is sig worthy. As it does seem that their monthly driver schedule has not helped them any over NVIDIA's schedule.
Originally posted by: SickBeast
The schedule of the drivers is not the issue; it's whether or not they work that counts.
There is no excuse for what's going on with FC2, but who's to say it's just ATi fault? What has Ubisoft done to correct the problem? What's Ubisoft done to fix widescreen support. Nothing nada.Originally posted by: nRollo
Read the rest of the page- Derek asks AMD to dump monthly releases, says he finds more problems with AMD drivers, and says there's no excuse for what is going on with FC2.
Yeah and then you get situations with NV where you can go 3 months or more without a single WHQL driver release, and betas that don't work properly.Originally posted by: Wreckage
Originally posted by: SickBeast
The schedule of the drivers is not the issue; it's whether or not they work that counts.
Rushing out drivers to meet a marketing schedule and then hotfixing em constantly is the issue.
Hell, why not have a daily driver release or an hourly driver release?
I think the Monthly release shtick is more a marketing ploy than an actual service to the customer. It's clear from recent issues these drivers don't have time to be properly tested and the consumer ends up being an unwilling beta tester.
Originally posted by: SickBeast
The schedule of the drivers is not the issue; it's whether or not they work that counts.Originally posted by: Wreckage
Maintaining a monthly driver release schedule is detrimental to AMD's ability to release quality drivers. This is not the first or only issue we've seen that could have been solved (or at least noticed) by expanded testing that isn't possible with such tight release deadlines.
Wow Apop just got served.
I would say that is sig worthy. As it does seem that their monthly driver schedule has not helped them any over NVIDIA's schedule.
I love how you and your brother Rollo mash down on apoppin about drivers now, and you were all quiet a year ago when NV could not produce a driver for their new 8800GTX cards in Vista. That situation was way way worse than what is going on now with AMD.
Pretty soon Rollo's kid will be old enough to post on here. Too much NV bias. :brokenheart:
So you're not denying that Wreckage is indeed your brother?Originally posted by: nRollo
Originally posted by: SickBeast
The schedule of the drivers is not the issue; it's whether or not they work that counts.Originally posted by: Wreckage
Maintaining a monthly driver release schedule is detrimental to AMD's ability to release quality drivers. This is not the first or only issue we've seen that could have been solved (or at least noticed) by expanded testing that isn't possible with such tight release deadlines.
Wow Apop just got served.
I would say that is sig worthy. As it does seem that their monthly driver schedule has not helped them any over NVIDIA's schedule.
I love how you and your brother Rollo mash down on apoppin about drivers now, and you were all quiet a year ago when NV could not produce a driver for their new 8800GTX cards in Vista. That situation was way way worse than what is going on now with AMD.
Pretty soon Rollo's kid will be old enough to post on here. Too much NV bias. :brokenheart:
I am schooling Rollo Jr in the dark side of the Force and forum ju jitsu....heh
I'm not mashing down on Apoppin, and wouldn't.
Just pointing out that Derek very specifically said ATi needs to address driver issues and that he said the monthly schedule hinders that.
Derek alluded to that in an earlier thread, but Apoppin didn't believe it, I guess because he wasn't as direct as this time.
I don't have anything against Apoppin, we just disagree on this issue.
So you're not denying that Wreckage is indeed your brother?[/quote]Originally posted by: SickBeast
I don't have anything against Apoppin, we just disagree on this issue.