PDA

View Full Version : Ars: ATI a takeover target?



UtwigMU
9th July 2005, 17:26
http://arstechnica.com/news.ars/post/20050709-5077.html


Takeover rumors started in June, when the company's third quarter results were less than impressive, and were followed by weak projections for the fourth quarter. Rumors indicate that both AMD and Intel are interested in ATI, as are Broadcom and Texas Instruments.

Gurm
9th July 2005, 18:28
http://arstechnica.com/news.ars/post/20050709-5077.html

NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!

Jammrock
9th July 2005, 19:23
NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO!
!!! [gasps for air] NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOO...

High_Jumbllama
9th July 2005, 20:17
:eek: :eek: :eek:

bsdgeek
9th July 2005, 21:20
TI could make some sick calculators. :crosseye:

UtwigMU
9th July 2005, 21:28
Well Creative took over 3D Labs and not much has changed for them.

Admiral
10th July 2005, 06:19
if it's true, I'm hoping for Intel :p

Taz
10th July 2005, 10:41
if it's true, I'm hoping for Intel :p

Yep then the Extreme in their extreme graphics controller might mean something positive :lick:

KvHagedorn
10th July 2005, 11:47
I hope they DON'T get taken over, but if they do, I would wish for AMD rather than Intel.

Wombat
10th July 2005, 11:55
There's a strong rumor, that I was never able to verify, that ATI had intel engineers come in and work on some of their chips, like the 8500. That's when ATI was fast as hell out of nowhere.

UtwigMU
10th July 2005, 12:32
There were rumours also about R9700 being hand tuned by intel engineers.

As per AMD, AMD is in bed with nVidia wrt to chipsets at the moment.

Liquid Snake
10th July 2005, 12:46
!!! [gasps for air] NOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO OOOOOO...


:eek: :eek: :eek:

I agree with these posts!

DGhost
10th July 2005, 13:28
**** that shit. they need a massive change of management. and their entire software department to get summarily fired.

****ing retards.

Taz
11th July 2005, 04:18
There's a strong rumor, that I was never able to verify, that ATI had intel engineers come in and work on some of their chips, like the 8500. That's when ATI was fast as hell out of nowhere.

If that's the case why haven't they produced a decent graphics controller of their own :confused:

Nowhere
11th July 2005, 04:26
Perhaps that wasn't Intels goal? Perhaps they didn't want Nvidia dominating the place?

GT98
11th July 2005, 06:02
**** that shit. they need a massive change of management. and their entire software department to get summarily fired.

****ing retards.

I thought I saw a rumor on Rage3D that ATI had their interns or 1st year graduates write their drivers!

That being said I haven't had problems with ATI drivers since the Orginal Radeon minus a couple goofs with the 8500 series

Jammrock
11th July 2005, 09:13
I can see AMD as the #1 buyer, because that would give them a foothold into the graphics market, where they don't currently have anything, but Intel does. This would also give them the ability to have their own motherboard chipsets for non-workstation/server applications, because that's the only market that AMD current makes chipsets for. A good offering in the OEM arena could open doors for AMD with certain businesses and OEMs *coughDELLcough* that have been hesitant about buying AMD because AMD doesn't make their own motherboards. That would be the ideal marriage in my book.

I can also Intel buying ATi to stop AMD from buying ATi and potentially getting a better foothold in the business and OEM world.

I can see braodcom becuase it would give them good expansion options and full line of computer peripherals.

I have no idea why TI is in the market. The best I could come up with for TI is maybe so they can get ahold of ATi's HDTV processor line to beef up their DLP line of HDTVs and video processors?

Jammrock

Nowhere
11th July 2005, 09:22
But can AMD afford to anger Nvidia? (best chipsets for a long time...and I don't see that'll change...)

Kurt
11th July 2005, 09:26
AMD to buy ATI would be a good business move, long term. Short term would be messy. They would have problems with NV (their N°1 chipset manufacturer) and ATi's chipset division is still battling for a bug-free chipset.

They would have, however, an excellent position against Intel in the mobile market, where ATi has excellent market shares and AMD is a laggard. That would be the single best reason for them to do it, IMO, since notebook sales are bound to grow above desktop sales in the near future.

Then again, AMD has much less resources than Intel and diversificatin might no be the best thing for them.

DGhost
11th July 2005, 09:37
TI does a lot of other businesses - they would aquire their Media products (Theater and Xilleon chips), their mobile/embedded market (Imageon products) (both of which could potentially augment their own markets), and then by buying into the video card/chipset business they expand their business portfolio.

Broadcom would probably push the Theater/Xilleon/Imageon chips far harder than the PC Graphics stuff too - their own preferred market is that as a chip/solution provider, not dealing with the end user.

both would win because of manufacturing experience, contracts and capabilities.

AMD might have a similar goal in mind. their embedded platforms division would gain a lot from those pieces. the graphics division would help round them out a lot, but I don't see the chipset department being a good investment for them. their own chipset, while lacking, does what they want it to. they never intended to make a living off of it - just to provide the platform for their processors. plus, to bring in ATI would possibly alienate their #1 ally right now - NVidia.

Intel has a lot to gain off of it. they would gain a lot of IP that would help bring their now dead graphics dept back to life. their engineers have a much better chipset going for them, but the graphics business would help them improve their integrated offering that much more. plus, being able to come up with a faster integrated solution will help their chipset adoption when Longhorn comes out. their other "little" businesses would benefit a lot from the other tech - the Imageon was designed to be able to work with their StrongARM processors and being able to merge it into the platform would secure their chipset as the platform of choice for a handheld app.

edit: IIRC, ATI's chipset biz came mostly from ALi/ULi. It is most definately not the best chipset out there, and with Intel's current deals with NVidia I could also see it as a way to strengthen this relationship. Intel can afford to kill it off and scrap the products for tech value only, and I could see NVidia very much appreciating that. with NVidia alreading offering an SLi compatable (and by all available information, Crossfire compatable) platform, Intel could be in quite the position to give NVidia their midrange desktop/lowend workstation chipset market - someplace that Intel has never really gotten right. the benefits they would get by graphics tech for their own integrated products would more than offset that.

Just remember, it's about who wants it the most, not who will do the most with it.

Fluff
11th July 2005, 09:52
How much money does Matrox have in the bank?

Gurm
11th July 2005, 09:59
How much money does Matrox have in the bank?

Not even enough to hire back the engineers that NVidia poached. :(

Gurm
11th July 2005, 10:13
I have remained relatively quiet in this thread, because some of what I might otherwise want to say might be a breach of NDA.

HOWEVER...

Disclaimer: I'm saying this based on personal conjecture only and in no way representing ATI:

As for the rumors that Intel helped ATI design the 8500/9700. Those are all clearly false for the following reasons:

1. Intel STILL doesn't have a chipset that is even as fast as the 8500, which was market leader in... 2001? If they had helped design the 8500 and/or 9700, why do their chipsets still suck donkey balls? In fact, if anything I would guess that Intel might have gotten occasional outside help. Once in a while they have non-sucking chipsets. For a while the i740 was a good chipset, and some of their integrated graphics now aren't ENTIRELY terrible. But ATI's bottom-of-the-barrel bargian chipset... outperforms everything Intel has for graphics.

2. ATI acquired several smaller ventures over the past few years. They acquired Oak Technologies, or at least part of it. Oak had acquired Xionics, and Xionics had some talented driver writers when I worked there. Oak, in general, has had very COMPETENT driver writers for their various chipsets.

-------------------------

On the topic of drivers...

ATI's drivers have consistently gotten BETTER with each iteration since the introduction of "Catalyst". Around the time that Neverwinter Nights came out (summer 2002) was the last time that ATI had SERIOUS driver issues that weren't promptly rectified. Within the next couple months, the Catalyst project began to crank out drivers, which were consistently MUCH BETTER than the previous drivers. And they've just gotten better.

While there are still a number of games from that era that DO have some issues (Half Life - the original - still has a few issues here and there with ATI's drivers), ever since they've done very well. They've taken a LOT of bashing that wasn't deserved. Drivers come out nearly monthly, and each time there are bugfixes and even speed improvements. Hell, new drivers still offer speed improvements for 8500-series cards!

NVidia, on the other hand, has NEVER gotten their driver act together. Their chipset drivers are so inconsistent that virtually everyone uses "remixes" with the safe versions of all the drivers rolled into one package. Each new NVidia video driver release brings with it the possibility of breaking some game that worked with the last release. And their support for their older cards is lacking - there's a bug in Call of Duty that causes some GF2 class cards to go all "yellow and black triangle-ish" that, last time I checked, was STILL THERE. Has been for a year now.

So honestly I don't know where the anti-ATI driver vitriol comes from. My GUESS is that it sort of trickles down from the gamer magazines and review sites, who are always a bit irked that ATI's OpenGL drivers are perpetually behind NVidia's in the numbers. When you glorify FPS above all else, you tend to be derisive about a company that doesn't make that their top priority. And that derisiveness filters down.

mmp121
11th July 2005, 10:34
How much money does Matrox have in the bank?

LOL, you beat me to the punch Fluff! I was gonna post something similar about Matrox takin em over.

DGhost
11th July 2005, 10:46
my Anti-ATI sentiment comes from the fact that every ATI card I've ever owned has given me problems until the day I got rid of it. And then it gave the people who got it problems. every single card.

The OpenGL drivers are not simply slow, they are *BAD*. when working on development projects things would break with every release of the drivers. not just small things either. to try to do any OpenGL development on an ATI display adaptor is equivilent to bashing your head into the wall repeatedly.

Gurm, the comments you have on ATI versus NV are bullshit. I can still ****up an 8500 just as fast as I could back in the day, and NV's driver updates give some performance increases to the GeForce3's as well. it may not be much, but it is easily on par with what ATI claims for their older architecture.

both companies have had their issues with support. I remember the ATI TV Wonder I purchased that constantly went from beta to beta drivers on XP. not that big of an issue, except for the fact that they would wait until 1 day shy of a year (literally) between release, and the driver set had more issues that a hacked/converted set of All in Wonder drivers. no other company had the same number of issues going from 2k to XP. and every other company had resolved their conversion issues by a year after, whereas ATI *still* had problems a year or two after the RTM date.

you say NV's products require constant hacks and repackages - last I checked it was ATI's drivers that started (and still continue) that trend. how many people over on Rage3d acctually run stock ATI drivers these days? Every time I poke over there to check things out I routinely see people recommend that they use the Omega drivers *instead of* the ATI released ones.

you say their drivers have gotten better... i have to agree with that in the "no longer causes random BSOD's on regular intervals" sense. last time I used an ATI card it still was responsible for causing BSOD's on a non-regular interval though. or their lovely VPU recover feature kicking in at a bad moment.

NV's driver set has been the standard for software development and stability, rivaled only by Intel. not only have I never seen a BSOD in either XP x86 or XP x64, the level of compatability and performance offered by their drivers under those products was something that I had not seen for many, many years. they had nice stable drivers up and running on an in development OS before ATI even had drivers that would BSOD in a repeatable fashion.

*shrug*

Tjalfe
11th July 2005, 11:01
As for the rumors that Intel helped ATI design the 8500/9700. Those are all clearly false for the following reasons...




I believe I recall reading that intel helped with implementing power saving and general speed optimizations, nothing actual architechtural.. but I could be wrong

Mikko
11th July 2005, 11:18
Intel STILL doesn't have a chipset that is even as fast as the 8500, which was market leader in... 2001? If they had helped design the 8500 and/or 9700, why do their chipsets still suck donkey balls?I suppose faster graphics would definitely have some impact on a price of the motherboards. And not all customers are willing to pay that price. Let's say you have a motherboard of $100. I think that maybe only 10% of the whole price comes off the graphic chip, at maximum. You don't get anything spectacular with this amount of money...

Gurm
11th July 2005, 12:48
my Anti-ATI sentiment comes from the fact that every ATI card I've ever owned has given me problems until the day I got rid of it. And then it gave the people who got it problems. every single card.

Sucks for you. What are you doing wrong? I'm not just being an ****ole here, but I've never... NEVER... had this kind of problem with an ATI card. None. I have with nVidia cards. :(


The OpenGL drivers are not simply slow, they are *BAD*. when working on development projects things would break with every release of the drivers. not just small things either. to try to do any OpenGL development on an ATI display adaptor is equivilent to bashing your head into the wall repeatedly.

You write "OpenGL development". Are we talking about games here? Or something more serious? It matters, you know. ATI sells a completely different line of cards to do anything other than game.


Gurm, the comments you have on ATI versus NV are bullshit. I can still ****up an 8500 just as fast as I could back in the day, and NV's driver updates give some performance increases to the GeForce3's as well. it may not be much, but it is easily on par with what ATI claims for their older architecture.

Wow, amazing. I've had EXACTLY the opposite experience. I can make a GF2 die a horrible, gasping death in about a minute... whereas I can't think of a single game/app that'll crash an 8500. Not one.


both companies have had their issues with support. I remember the ATI TV Wonder I purchased that constantly went from beta to beta drivers on XP. not that big of an issue, except for the fact that they would wait until 1 day shy of a year (literally) between release, and the driver set had more issues that a hacked/converted set of All in Wonder drivers. no other company had the same number of issues going from 2k to XP. and every other company had resolved their conversion issues by a year after, whereas ATI *still* had problems a year or two after the RTM date.

Oh, you're just not thinking, now. I had a dozen pieces of hardware that just had... ISSUES going to XP. My video capture card... unusable under XP. My MIDI board? Unusable. Creative's sound cards on my SMP system? Unusable under 2k OR XP. C'mon now. You're just being angry because you had some retrograde motherboard that didn't like ATI cards. ;)


you say NV's products require constant hacks and repackages - last I checked it was ATI's drivers that started (and still continue) that trend. how many people over on Rage3d acctually run stock ATI drivers these days? Every time I poke over there to check things out I routinely see people recommend that they use the Omega drivers *instead of* the ATI released ones.

When's the last time you checked? Omega hasn't done a repack in over a year. This is what I mean. You're basing your entire estimation of their driver support on the way their driver support WAS. In 2003 I might have agreed with you.


you say their drivers have gotten better... i have to agree with that in the "no longer causes random BSOD's on regular intervals" sense. last time I used an ATI card it still was responsible for causing BSOD's on a non-regular interval though. or their lovely VPU recover feature kicking in at a bad moment.

You know what? I've never seen a BSOD that wasn't from something else. The only ATI-related BSOD I've seen was an SMP-related issue on my VIA 694D that was rectified with a MOTHERBOARD BIOS update.


NV's driver set has been the standard for software development and stability, rivaled only by Intel. not only have I never seen a BSOD in either XP x86 or XP x64, the level of compatability and performance offered by their drivers under those products was something that I had not seen for many, many years. they had nice stable drivers up and running on an in development OS before ATI even had drivers that would BSOD in a repeatable fashion.

Really? News to me. I'm not talking about BSOD's, though. It's not excusable for ANYONE to have those. I'm talking about apps that crash, hang the machine, or just look like crap. And for you to deny that nVidia breaks their drivers every other build is ... simply perpetuating a monstrous falsehood. Every time I log into the nVidia forum, the advice is "well drop back 3 driver sets, that game worked two months back..."


*shrug*

See, I think the problem here is that you must have had some sort of horrible motherboard. Maybe using a VIA chipset, hmm?

Gurm
11th July 2005, 12:49
I suppose faster graphics would definitely have some impact on a price of the motherboards. And not all customers are willing to pay that price. Let's say you have a motherboard of $100. I think that maybe only 10% of the whole price comes off the graphic chip, at maximum. You don't get anything spectacular with this amount of money...

ATI's motherboards have R3x0 cores onboard. And they're cheaper than Intel's boards. nForce boards have GF2 graphics on the early ones, GF3 on the intermediate ones, and I imagine GF4 or better on the brandy-spanking new ones. THOSE boards are also cheaper than Intel's.

DGhost
11th July 2005, 13:14
LOL. VIA chipsets for some. NV for others. Intel on some of the others. If I were to reach I can even remember some AMD chipset systems running ATI.

RE: OpenGL dev was for games. thankfully, I was not the one working on the ATI code path. yes, the team was in constant contact with ATI's dev team who was pretty friendly with it. they would get something working, document the problem, send it to them who would thank us, suggest a workaround, and then tell us it would be fixed in the next driver release. we would wait until the next driver release and when they would bring it up and running on the new driver set the specific problem would be fixed, but a new problem would exist. wash, rinse, repeat. some of these were fairly specific code problems.

anyways, you mention the magic number of 2003. yes, most of this took place in 2003. however, my friends were the poor souls who tried 9600XT cards and were woefully dismayed by what it did to them. one was on an Intel chipset (i875/ICH5R, iirc) and I am not sure about the other. one ditched it in favor of a 6600GT when they became available and has been 3x as happy ever sense. the other still is limping along with his (I believe 9600Pro) and is subject to an insane amount of issues. he still gets cranky about it.

RE: XP. I was on the beta team for XP. there were a lot of pieces of hardware that were not supported, and a lot that were. i suppose most of my buying of hardware was probably more influenced by what *did* while it was still in beta. yes, a lot of hardware had issues. NV's drivers were insanely superior at the time. they still are. it took ATI forever and a ****ing day to get true dual-head implemented in a manner that didn't completely hose the system. and the TV Wonder... I had at least another TV tuner with the same chipset that had full XP support. for most hardware it was not a big deal to convert drivers, especially if they already had 2k driver out there (which ATI *did*). I believe I was using an 8500 right around then, although I had a GeForce 256 and a G400-Max that I liked to play with as well.

about the BSOD's... when you get a BSOD in atixxxxx.dll or .sys it's a pretty good indicator of who or what is at fault. and just about every time I've seen a computer doing that, I can drop an NV card in and it will work without issue. now... maybe thats an issue with the MB... maybe thats an issue with the video card... simple fact of the matter is that ATI didn't work and NV did...

as far as specific rendering issues... i've seen them on ATI. I saw them all the time, especially (although, probably mostly) under OpenGL. dealing with Quake3 graphics issues in 2003 is a little bit old, don'tcha think?

Jon P. Inghram
11th July 2005, 13:17
OMG, did you have to pledge undying loyalty to ATI to become a beta tester or what!? Feels like we're being invaded by fanATIcs from Rage3D... :rolleyes:

Mikko
11th July 2005, 13:39
ATI's motherboards have R3x0 cores onboard. And they're cheaper than Intel's boards. nForce boards have GF2 graphics on the early ones, GF3 on the intermediate ones, and I imagine GF4 or better on the brandy-spanking new ones. THOSE boards are also cheaper than Intel's.I believe that GF2, GF3 and GF4MX will beat Radeon 8500 in very rare situations only.

Gurm
11th July 2005, 13:40
No, just eternal hatred of nVidia.

Seriously, though I'm not especially biased. I have had my share of GeForce cards. I'm STILL biased against them on the basis of image quality, although admittedly they've gotten quite a bit better. Nobody will ever stack up to Matrox in that arena though. I still have a G550 I use in one machine, I just don't play games on it.

But I am a bit leery of nVidia. I've seen too many GeForce cards MELT. I've never seen an ATI card blow its caps.

There have been debacles, sure. The early 9600's were problem-ridden. No question. That was a huge issue. ATI also have this habit of dropping not-quite-ready-for-primetime "features" into their release drivers. The automatic overclocking wasn't QUITE ready. Neither was the exception recovery. Neither was CCC, to be honest.

But Christ, there are new nVidia drivers all the damn time. From nVidia, from other sources, betas here, release drivers there. Can't bloody keep TRACK of them. On the other hand, you can be PRETTY SURE that Catalyst 3.4 is, in fact, both newer and better than Catalyst 3.3 was. You can't say that with ForceWare 54.77 versus Forceware 53.12, half the time the 53.12 is NEWER, or the 54.77 has some horribly broken component.

And nVidia has been CONSISTENTLY bad. Their OpenGL has always been GREAT. I hand them that. No question, if you want the latest game from ID to run flawlessly, plug in an nVidia card. When RTCW came out, I played it exclusively on a GF3ti200, my 8500 collected dust. But everything not OpenGL related has been subject to extreme clusterfvcking from driver to driver.

Also there's the heat thing. And the EXTENSIVE "cheating on benchmarks" thing. And the engineer poaching thing. And the "crushing 3dfx because they had a better product" thing. Overall, NVidia is just a scummy company. That's not my professional opinion, that's my PERSONAL opinion.

I admit that they're faster on a regular basis. Their OpenGL is better on a regular basis. But the rest of it stands.

Gurm
11th July 2005, 13:42
I believe that GF2, GF3 and GF4MX will beat Radeon 8500 in very rare situations only.

Or in any OpenGL game. Let's call a spade a spade...

leech
11th July 2005, 20:08
There is only ONE reason I went with nVidia over ATI; Linux.

ATI's drivers still consistently suck under linux. From weird redraw errors on the desktop, to OpenGL games being completely hosed.

nVidia's drivers have their share of issues as well, but they at least work and work well under linux.

Leech

Victorian
12th July 2005, 01:07
When's the last time you checked? Omega hasn't done a repack in over a year. This is what I mean. You're basing your entire estimation of their driver support on the way their driver support WAS. In 2003 I might have agreed with you.


Huh? He's still doing his thing. Just checked his website and he's current with the 5.6's

The PIT
12th July 2005, 02:53
There is only ONE reason I went with nVidia over ATI; Linux.

ATI's drivers still consistently suck under linux. From weird redraw errors on the desktop, to OpenGL games being completely hosed.

nVidia's drivers have their share of issues as well, but they at least work and work well under linux.

Leech

And the latest ATI drivers still won't work for me under Suse Linux 9.3.

At least I've got the system working at 1600 x 1200 without the ATI drivers and since I'm not gaming it can stay like that for now.

Kurt
12th July 2005, 03:28
and I'm using the latest and greatest ATi driver on a PC (thank god not my main one anymore!) and dual head SVCKS majorly. OK, it's not a bug, just a feature.

they still haven't figured out how to make it work without making you lose _more_ time than with only one screen.

no games on that PC so I'm not complaining about them. haven't had much trouble with ATi's in that respect, except with my old Rage128.

most trouble I had with BSOD was with NV Ti4600 on win2k. that has passed. now I have problems with my win2k install on Nforce4. not sure it's them, but there isn't much on it (not even office) and it does strange things at times (like mouse problems opening windows and things - no spyware or other nasties whatsoever, could be a bad KB/mouse not sure)

Gurm
12th July 2005, 06:15
Huh? He's still doing his thing. Just checked his website and he's current with the 5.6's

Look at them. They aren't a repack. They're a bundle - Catalysts + Tweaker + Presets. No mixing and matching. No driver twiddling. No "OpenGL from a previous build".

UtwigMU
12th July 2005, 17:14
Well currently based on few Ati owner's experience and on my experience Matrox Parhelia Linux drivers are better than Ati. Also framerate is on par with Windows in Enemy Territory (Parhelia).

Otherwise I've seen a lot more Ati 9xxx issues than nVidia 5xxx and 6xxx issues on Arstechnica forums and saying that everyone had 9xxx is not true as there were a lot of people either holding to their Ti 4's or even got FX series because of being fanboys.

Mehen
12th July 2005, 19:30
ive had some issues with 5xxx cards, some of the driver updates did stupid things, there was about 4 months worth of drivers that came out (early 6x.xx's i think) where my tv-out didnt work.
as for ATI cards, my brother hasnt ever had any issues with his 9800xt, and ive never had any issues with the 7000's ive had, and ive never heard anything bad about the 9600pro/xts that some of my friends have.
On the other hand, the 128mb 8500le that we had in the house for awhile gave us nothing but problems...

Gurm
13th July 2005, 07:00
ive had some issues with 5xxx cards, some of the driver updates did stupid things, there was about 4 months worth of drivers that came out (early 6x.xx's i think) where my tv-out didnt work.
as for ATI cards, my brother hasnt ever had any issues with his 9800xt, and ive never had any issues with the 7000's ive had, and ive never heard anything bad about the 9600pro/xts that some of my friends have.
On the other hand, the 128mb 8500le that we had in the house for awhile gave us nothing but problems...
There were certain cards that should never have been sold. The 8500LE was never properly equipped to handle 128MB. I'm not certain that the 8500 core was set up to handle that much memory AT ALL, but I have no proof one way or the other, no inside knowledge that would lead me to conclude that. The only people that have R2x0 cores with 128MB that have been VERY HAPPY are those with much newer ones (9250, for example) or a few people with JUST THE RIGHT BIOS on an old 8500/128 DDR. But even those were sort of hacked-together cards. The BEST 8500 was the 8500/64/DDR "Built-By-ATI".

Nowhere
13th July 2005, 07:07
It's always nice to convince yourself that you have the best, huh? ;) :p

Kurt
13th July 2005, 07:22
@|Mehen|: I've had tons of problems with the 7500 cards and early 7000/VE. They were drivers issues mainly. 7200 and 8500 were working fine and I never had a problem with my 9700 or 9800pro.

Gurm
13th July 2005, 07:50
It's always nice to convince yourself that you have the best, huh? ;) :p

No, I actually went out and BOUGHT one because it was the best-of-series.

Gurm
13th July 2005, 07:51
@|Mehen|: I've had tons of problems with the 7500 cards and early 7000/VE. They were drivers issues mainly. 7200 and 8500 were working fine and I never had a problem with my 9700 or 9800pro.

Yep. The 7000/7500, known at the time primarily as "Radeon DDR" or "Radeon SDR", were plagued with problems. This was the six-month period before ATI switched to "Catalyst" drivers, the old drivers were HUGELY problematic. ATI got a big influx of driver talent from SOMEWHERE around that time.

Chrono_Wanderer
13th July 2005, 07:54
I thought I saw a rumor on Rage3D that ATI had their interns or 1st year graduates write their drivers!

That being said I haven't had problems with ATI drivers since the Orginal Radeon minus a couple goofs with the 8500 series

eh. there are first year computer engineering students in my class working at ATI right now. i think its mostly drivers stuff. (probably QA, not core development)

Kurt
13th July 2005, 08:00
ATi's very own Waterloo... :rolleyes:

:D

The PIT
13th July 2005, 08:29
Guess thats why there Linux drivers are so shite. The students are used to windoze.

Liquid Snake
13th July 2005, 12:30
Yep. The 7000/7500, known at the time primarily as "Radeon DDR" or "Radeon SDR", were plagued with problems. This was the six-month period before ATI switched to "Catalyst" drivers, the old drivers were HUGELY problematic. ATI got a big influx of driver talent from SOMEWHERE around that time.

Yeah I had a 7500 for a while. The pre-Catalyst drivers had some problems but nothing that trashed my system. Afterwards I had no problems, and I have no problems with my current 9800 Pro either.

Mehen
13th July 2005, 19:42
ahhhh, well most of the experience ive had with the 7000 series has been in the last 2 years, mostly people just buying the cheapest card they can, that would explain why they have always seemed alright to me

Chrono_Wanderer
13th July 2005, 21:43
hack, believe it or not, my first year programming course was C# on Windows 2000 :rolleyes:

Kurt
14th July 2005, 00:30
Yeah I had a 7500 for a while. The pre-Catalyst drivers had some problems but nothing that trashed my system. Afterwards I had no problems, and I have no problems with my current 9800 Pro either.

I couldn't even change the driver from the RELEASE one for most of the cards. Any other driver would mess up the display exactly like a faulty RAM chip would have. Reinstall the original driver and the problem was fixed. Luckily the new drivers fixed that and the cards' drivers were updated and all's well.

This should have NEVER happened.

Gurm
14th July 2005, 04:15
I couldn't even change the driver from the RELEASE one for most of the cards. Any other driver would mess up the display exactly like a faulty RAM chip would have. Reinstall the original driver and the problem was fixed. Luckily the new drivers fixed that and the cards' drivers were updated and all's well.

This should have NEVER happened.

Happens all the time, to every company. nVidia does this every second or third release for one or more of their cards.

Kurt
14th July 2005, 04:29
That's the problem. It SHOULD NOT happen, EVER. Companies have it too easy these day with the web and an easy way to update/upgrade software. Makes for sloppy releases that "can be patched" later...

Of course, this never happening is pure utopianism...