What keeps you going to Intel?

[santaliqueur]

I pretty much always use intels for my clients. The reason, less problems with compatability and general use. I hate when I just build a system and I get a call and somthing is not working correctly. Ive opnly had this on AMD systems. Intel systems have been great and I continue to use them. Less time means more money for me.

Noone said it was the processor

i'm pretty sure that is what scotch77 is saying here.

Intel should buy AMD, imagine the pair, good awsome buisness meets great developement.

that enormous sucking sound you hear is the competitive nature of intel being sucked away once they bought amd. we are benefitting from the competition, intel buying amd would be the worst thing that could happen for us.

dualdrop[H]enia

 

[Dan_D]

i'm pretty sure that is what scotch77 is saying here.


that enormous sucking sound you hear is the competitive nature of intel being sucked away once they bought amd. we are benefitting from the competition, intel buying amd would be the worst thing that could happen for us.

dualdrop[H]enia

Agreed.

 

[robberbaron]

Back in the T-Bird and earlier days I would have agreed with the posters statement about AMD having a higher failure rate. These days, I would not agree. I've built many A64 based systems, and I have yet to ever see a DOA S754, S939, or S940 chip.

In 9 years of being a tech, I've never seen a DOA Intel processor. Ever. Overall I can not say the same for AMD.

In my 4 months of being a tech (professionally) I've seen two DOA P-D 820's and two DOA Sempron 2800+'s.

 

[Dan_D]

In my 4 months of being a tech (professionally) I've seen two DOA P-D 820's and two DOA Sempron 2800+'s.

Honestly, I have yet to actually unbox a Pentium-D. I can't speak to those. It wouldn't surprise me if the failure rate was higher on dual core CPU's. It would make sense.

I've never seen a dead dual core Pentium-D in a new OEM machine though that was taken right out of the box. Not yet anyway. Seen plenty of dead Semprons though.

 

[ALL4AMD]

i forget which page it was on but someone mintioned that the IHS increases the contact area to the HSF but thats stupid... it doesn't increase the cores contact area to the IHS.

i have to aggree its chipset issues that cause "stability issues"

back to the slot 1/A issue.... The reason they did off die l2 cash is because the cpu die would be too big and too expensive @ .25 ever c a .25 thunderbird no because they were .18?

and to people that say the thunderbird ran hotter than hell.... no they don't run hotter than hell they just put out more heat @ the time than intel but look now.... both amd and intel put out more heat than the thunderbird or about the same

 

[contoursvt]

and to people that say the thunderbird ran hotter than hell.... no they don't run hotter than hell they just put out more heat @ the time than intel but look now.... both amd and intel put out more heat than the thunderbird or about the same

Great, so maybe someone can release a CPU tomorrow which puts out 500W of heat into a 1cm x 1cm area and we'll pretend its not running hot because 3 years from now, maybe all cpus will be running that hot and giant liquid nitrogen coolers will be the norm

The point is that when the tbirds came out, they were way WAY hotter than anything else so there was a cooling issue with them using normal coolers. People had to resort to copper coolers with high speed fans because at the time, nothing else was even close to putting that much wattage through that small a die.

 

[ALL4AMD]

The point is that when the tbirds came out, they were way WAY hotter than anything else so there was a cooling issue with them using normal coolers. People had to resort to copper coolers with high speed fans because at the time, nothing else was even close to putting that much wattage through that small a die.

but u have to remember coolers back then were the size of todays northbridge cooler.

i beg to differ on that nothing has tried to put out that much heat but yet the Tbred XPs were smaller and put out about the same amount of heat

 

[chrisf6969]

Great, so maybe someone can release a CPU tomorrow which puts out 500W of heat into a 1cm x 1cm area and we'll pretend its not running hot because 3 years from now, maybe all cpus will be running that hot and giant liquid nitrogen coolers will be the norm

The point is that when the tbirds came out, they were way WAY hotter than anything else so there was a cooling issue with them using normal coolers. People had to resort to copper coolers with high speed fans because at the time, nothing else was even close to putting that much wattage through that small a die.

I remember having a "high end" cooler back then was a fancy slot 1 aluminum HS with 2 fans on it! Nobody was using copper! http://www.gearxs.com/gearxs/product_info.php?products_id=4580

 

[contoursvt]

I remember having a "high end" cooler back then was a fancy slot 1 aluminum HS with 2 fans on it! Nobody was using copper! http://www.gearxs.com/gearxs/product_info.php?products_id=4580

Ummm ya nobody was using copper . Wake up dude, I had an alpha heatsink with a nice copper base which makes that joke of a heatsink you posted as high performace look like something that came out of a cereal box. Obviously you and I have a different definition of what a good heatsink is. Got my PIII 450 to run at nearly 600Mhz and remain ice cold.

http://www.hardware.fr/art/imprimer/171/

 

[chrisf6969]

Ummm ya nobody was using copper . Wake up dude, I had an alpha heatsink with a nice copper base which makes that joke of a heatsink you posted as high performace look like something that came out of a cereal box. Obviously you and I have a different definition of what a good heatsink is. Got my PIII 450 to run at nearly 600Mhz and remain ice cold.

http://www.hardware.fr/art/imprimer/171/

Ok SOME people had them... but most people back then were impressed my a DUAL FAN slot 1 cooler. I thought I was big pimpin with a similar one as the link I posted. The one I had looked slightly better. Anyway.... back to porn surfing. later

 

[ALL4AMD]

don't worry chris just look @ his post count then urs

 

[robberbaron]

A Slot-1 cooler pissing contest!


How I love this forum!

 

[1c3d0g]

Back to your regular programming!

 

[contoursvt]

don't worry chris just look @ his post count then urs

Some of us actually post on other forums and have lives and hobbies which take us away from the computer - LOL

 

[WesM63]

I guess it really boils down to what you want in a system. The age old saying would be multimedia = Intel, Gaming = AMD. In most cases this is still true, i am a loyal AMD Fan. I've only ever owned one Intel system i used as a daily rig, Dual LV 1.6ghz Xeons and boy did that thing scream. Sadly enough I sold it after having it for ~6months. I've just never needed that kind of system for what I do here at home. Gaming, Disributed Computing and browsing the net.

Till Intel can give me something worth buying in my mind, its AMD for me.


(ya ya.. here comes the Donthan remarks. Give me PCI-E and i'll think about it.)

 

[ALL4AMD]

Some of us actually post on other forums and have lives and hobbies which take us away from the computer - LOL

i do 2 but what i was implying was ur title and how long u have been here

 

[contoursvt]

i do 2 but what i was implying was ur title and how long u have been here

I know Thats why I said some of us post on other forums. I spend much more time on Ars and 2cpu and some others. Its much more busy there usually. This is a nice forum but its not my mainstay.

 

[BlindedByScience]

Some of us actually post on other forums and have lives and hobbies which take us away from the computer - LOL

So meaning that he just joint the forum and therefore knows nutting? an idology rather stupid if you ask anyone. And by the way, you did not refer to "how long been here" but post count. Does having 1K+ posts count as highly knowledgeable?

contoursvt and Nasgul - if you've got something releavant to the thread to contribute, fine but if not....don't post. You'd both do well to read and follow the rules you agreed to when you registered. Keep it on topic or threads get locked, and repeat offenders get "vacations". Pretty simple.....

Back on topic, please - B.B.S.

 

[FlatLine84]

i'm pretty sure that is what scotch77 is saying here.


that enormous sucking sound you hear is the competitive nature of intel being sucked away once they bought amd. we are benefitting from the competition, intel buying amd would be the worst thing that could happen for us.

dualdrop[H]enia

I definately agree no doubt, but could you imagine AMD if it was a bugger buisness, how inexpensive their procs would be, and how they would be able to usher things out better....

 

[Donnie27]

I have seen dead procs from both. Everything from Celeron 300A's to Xeon 3.6's...and yes, I have seen a fair amount of Socket A's die as well.

I couldn't disagree more but I'm not in any of position to have built as many computers either. From the hundreds I've seen though and those of at least 8 friends who constantly upgrade, I, nor they have ever gotten a DOA Intel Processor of any kind. The few that have died didn't just die for NO reason like the many dead AMD processors we've all had. I've had 15 AMD systems that go back to at least the 386 days and 9 Intel systems.

The dead Intel Processors I've seen have been caused by overclockers using more than 1.7 volts on P4s even as Intel warned them not to.

Motherboard and or Power Supply failure taking the Processor and other parts with it.

I've repaired Intel systems with Failed Fans and even heat sinks that snapped/popped off completely. The processor still worked and seemed unaffected.

A lady is still using a Pentium 100 I sold her in 1996, I still have the Pentium 200 that replaced that rig. I still have the P3 550 that I used to backup all of my files through my Crashlon days. I'm sorry, I'll never agree that with you guys that Intel Processors die just as much as AMD's. I've ripped out far too many dead AMD procs to make it even in the same ballpark. She (same lady above) keeps buying Intel processors for the same reason many folks buy Honda Cars.

AMD's current processors had problems as well and even the Revision E fixed some. When I complained about the all 4 slots full Glitch, folks on this thread said it was something I made up.

that enormous sucking sound you hear is the competitive nature of intel being sucked away once they bought amd. we are benefitting from the competition, intel buying amd would be the worst thing that could happen for us.

To buy AMD, first Intel would need to payoff the Germans and the EU

But something AMD followers don't seem to understand, competition is a two-way street and it damned sure is not just white hat wearing good old guy AMD saving the computer industry from big ole bad Intel. Intel is the reason X2 3800+ is selling for $310 (an offer I was made) instead of its peak price of $410.

Donnie27

 

[DejaWiz]

I just had to add more to this thread.


As many of you know, I'm currently a huge AMD fan. Does this mean I'm an intel hater? No, of course not! intel makes good processors. I just happen to feel that AMD makes great processors in comparison.

I use a P4 in my work system (not my choice, but meh) and a P-M in my laptop (which is my primary choice for mobile computing right now. Even at only 1.6GHz, that thing screams!). I have a backup system that my brother uses as his daily use/work from home when he feels like calling in system containing dual PIII 1000's. I have never been more impressed with a dual cpu system than that one. It keeps on chugging along and has no problems whatsoever with any of the latest software.

I suppose the main reason I don't use intel processors in my home pc right now is because of the higher cost to a comparative performing AMD processor for what I use my system for. This could definitely change in the near future.

With 65nm parts coming into focus from both companies, I think we will be seeing another competetive innovation war. Whichever platform has the best performance/price ratio is what I will go with. That is after considering any possible issues that may have to be worked out in future revisions - I'm not an early adopter by any means, as I prefer to play it safe when I spend my hard earned money on computer components. Heck, it took the Venice core to finally get me to retire my Athlon XP and upgrade to an A64.

 

[Josh_B]

I couldn't disagree more but I'm not in any of position to have built as many computers either. From the hundreds I've seen though and those of at least 8 friends who constantly upgrade, I, nor they have ever gotten a DOA Intel Processor of any kind. The few that have died didn't just die for NO reason like the many dead AMD processors we've all had. I've had 15 AMD systems that go back to at least the 386 days and 9 Intel systems.

The dead Intel Processors I've seen have been caused by overclockers using more than 1.7 volts on P4s even as Intel warned them not to.

Motherboard and or Power Supply failure taking the Processor and other parts with it.

I've repaired Intel systems with Failed Fans and even heat sinks that snapped/popped off completely. The processor still worked and seemed unaffected.

A lady is still using a Pentium 100 I sold her in 1996, I still have the Pentium 200 that replaced that rig. I still have the P3 550 that I used to backup all of my files through my Crashlon days. I'm sorry, I'll never agree that with you guys that Intel Processors die just as much as AMD's. I've ripped out far too many dead AMD procs to make it even in the same ballpark. She (same lady above) keeps buying Intel processors for the same reason many folks buy Honda Cars.

AMD's current processors had problems as well and even the Revision E fixed some. When I complained about the all 4 slots full Glitch, folks on this thread said it was something I made up.



To buy AMD, first Intel would need to payoff the Germans and the EU

But something AMD followers don't seem to understand, competition is a two-way street and it damned sure is not just white hat wearing good old guy AMD saving the computer industry from big ole bad Intel. Intel is the reason X2 3800+ is selling for $310 (an offer I was made) instead of its peak price of $410.

Donnie27

So basically, everything Intel has ever put out is great? What about the PIII 1.13GHz, i820 w/MTH, etc etc.

Everyone forgets about those, but they damn AMD when they make the smallest mistake.

Additionally, anecdotal evidence is generally considered invalid. We need some hard figures on failure rates to actually come to any real conclusion.

 

[gerbiaNem]

I definately agree no doubt, but could you imagine AMD if it was a bugger buisness, how inexpensive their procs would be, and how they would be able to usher things out better....

Or they could get lazy like Intel from being on top so long...

 

[Chris_Morley]

Donnie27-

If you don't believe me it's all very well documented internally at BOXX Technologies. In fact, before I left I remember there being TRAYS of Intel procs waiting to be RMA'd. No overclocks whatsoever. There were plenty of AMD chips too, but don't be surprised that Intel procs fail.

 

[Dan_D]

So basically, everything Intel has ever put out is great? What about the PIII 1.13GHz, i820 w/MTH, etc etc.

Everyone forgets about those, but they damn AMD when they make the smallest mistake.

Additionally, anecdotal evidence is generally considered invalid. We need some hard figures on failure rates to actually come to any real conclusion.

No one argues this point. Or they shouldn't. AMD and Intel make mistakes. Intel is usually really quick to own up to them, and they get them fixed quickly. The 1.13GHz Pentium III, the 820 MTH chip, and let's not ever forget the Pentium 60 and 66 with the FPU bug.

Donnie27-

If you don't believe me it's all very well documented internally at BOXX Technologies. In fact, before I left I remember there being TRAYS of Intel procs waiting to be RMA'd. No overclocks whatsoever. There were plenty of AMD chips too, but don't be surprised that Intel procs fail.

I've built hundreds of machines and serviced thousands of others. It doesn't surprise me. Percentage wise though I'd have to say in my experience, I've seen more failed AMD cpu's than Intel's. DOA or otherwise. Even if I did see more Intels itt still doesn't mean anything, because Intel puts out alot more CPU's per year than AMD does. So naturally in sheer numbers alone their will be more dead Intel CPU's from failures or just being bad out of the box than AMD will.

What always used to keep me going back to Intel, was the total stability and quality of the platform vs. AMD based systems and solutions. That's changed over the last couple years. Especially since the introduction of the A64 and the chipsets that go with it.

 

[Donnie27]

So basically, everything Intel has ever put out is great? What about the PIII 1.13GHz, i820 w/MTH, etc etc.

Everyone forgets about those, but they damn AMD when they make the smallest mistake.

Additionally, anecdotal evidence is generally considered invalid. We need some hard figures on failure rates to actually come to any real conclusion.

Hell no I don't think everything Intel made or makes is great.

Oh, I remember it well. I also remember that Intel only shipped about 200 1133MHz Processors. Unlike others companies, Intel did recalled and tried to correct their problems.
The first Tbred should have been recalled as well as the first Prescotts. So Ask Kyle how long it took Intel to recall after the Problem was recreated? AMD, VIA and nVidia had equally or even more screwed up hardware that should have been recalled. They didn't do a damned thing about it.

AMD didn't make a "small mistake" it was just plain mistakes and BS=P The problem was and still is, they just don't ever admit it. Neither do their Fans. It's not a mistake to do what AMD did as they bragged about 64bit and how much RAM it addressed when installing with all slots full limited the REAL TOTAL and had problems with even 4GB, sheesh! That's the kind of Sleeaze Intel normally pulls with its crappy Gaming ads.

Anecdotal evidence is valid to each person buying whatever hardware=P Message boards use to be about getting good info, not about how some Fan love AMD or Intel.

Donnie27

 

[Donnie27]

Donnie27-

If you don't believe me it's all very well documented internally at BOXX Technologies. In fact, before I left I remember there being TRAYS of Intel procs waiting to be RMA'd. No overclocks whatsoever. There were plenty of AMD chips too, but don't be surprised that Intel procs fail.

I have no reason to doubt you. Maybe problematic Processors were sent back by OEMs and VARs before I got my grubby hands on them, but then why did they miss the AMD processors?

I've just not had Intel Processors die like some of my AMD processors or those yanked from other systems. In fact, I've not heard of my buds having Hammers die like the K6 (all), AthlonTB-XPs and or Classics. I waited for Rev-E and that's what I'll use for my next build. I didn't like the first nForce3 150 that was slower than a VIA K8 board and 250/GB had a short lifespan as well.

Donnie27

 

[M3at]

I have an Intel 2.8 with 800fsb. The applications I run on it are, Photoshop, VisualStudio, Enterprise manager, query analyzer and FrontPage 2003. Often these apps are all open all day long at the same time along with windows media player and a bunch of browsers.

I have to say, I see no difference in with HyperThreading turned on or off. In fact, I've had more crashes and more "laggy" problems with HyperThreading on as opposed to having it turned off. So, I don't use HyperThreading anymore.

I doubt, based off of my experience, that I will buy a current Intel cpu solely for HyperThreading as I see no real benefit, again, based off of my experience. If I buy Intel, it will be because the price and performance are what I need; however, AMD's prices and their performance just can't be beat especially if you are building a new system.

I remember my first AMD build. It was a t-bird 1333. My system sounded like an international airport. My next few builds after that were with the AMD XP series of chips (probably built a dozen 2500+XP’s). These chips did not run as hot and as a result my systems were quieter. I've had a water cooled system as well with the only fans being the one on my vid card and the 120mm on my radiator and of course psu fans. This system was virtually silent. I then sold the water cooled system and went for my first s754 build. This system has one 120mm for the case, vid card fan and cpu fan and still, it's barely audible. I was amazed at how silent it was and how low the temps were. With current AMD chips, I see no real benefit to water as a solution for silence. These chips run very cool on air, now if you are overclocking, that’s a different story. I have built around 20 a64 systems this year, both s754 and s939, for friends and family and I have not had one person come back to me because of instability problems or because of general system problems.

 

[Dan_D]

I have an Intel 2.8 with 800fsb. The applications I run on it are, Photoshop, VisualStudio, Enterprise manager, query analyzer and FrontPage 2003. Often these apps are all open all day long at the same time along with windows media player and a bunch of browsers.

I have to say, I see no difference in with HyperThreading turned on or off. In fact, I've had more crashes and more "laggy" problems with HyperThreading on as opposed to having it turned off. So, I don't use HyperThreading anymore.

I doubt, based off of my experience, that I will buy a current Intel cpu solely for HyperThreading as I see no real benefit, again, based off of my experience. If I buy Intel, it will be because the price and performance are what I need; however, AMD's prices and their performance just can't be beat especially if you are building a new system.

I remember my first AMD build. It was a t-bird 1333. My system sounded like an international airport. My next few builds after that were with the AMD XP series of chips (probably built a dozen 2500+XP’s). These chips did not run as hot and as a result my systems were quieter. I've had a water cooled system as well with the only fans being the one on my vid card and the 120mm on my radiator and of course psu fans. This system was virtually silent. I then sold the water cooled system and went for my first s754 build. This system has one 120mm for the case, vid card fan and cpu fan and still, it's barely audible. I was amazed at how silent it was and how low the temps were. With current AMD chips, I see no real benefit to water as a solution for silence. These chips run very cool on air, now if you are overclocking, that’s a different story. I have built around 20 a64 systems this year, both s754 and s939, for friends and family and I have not had one person come back to me because of instability problems or because of general system problems.

Totally can't agree on the Hyperthreading thing. On the comment about S754 and S939 AMD chips, I agree. As long as you build the computer with good components, either AMD or Intel chips will suffice in terms of quality. Both make a very good, and stable CPU.

 

[Tanky]

Totally can't agree on the Hyperthreading thing. On the comment about S754 and S939 AMD chips, I agree. As long as you build the computer with good components, either AMD or Intel chips will suffice in terms of quality. Both make a very good, and stable CPU.

The appropiate thing to say here, is that component wise, often times, you get what you pay for. Generally, the better quality components you start out with, the less problems you will have down the road. Intel chipsets and nForce4 chipsets come to mind.

 

[Donnie27]

The appropiate thing to say here, is that component wise, often times, you get what you pay for. Generally, the better quality components you start out with, the less problems you will have down the road. Intel chipsets and nForce4 chipsets come to mind.

I never had any problems with X-Fi, a few of my nForce4 friends had to wait on BIOS updates. I've just had fewer compat problems with all kinds of cheap or good stuff.

Totally can't agree on the Hyperthreading thing. On the comment about S754 and S939 AMD chips, I agree. As long as you build the computer with good components, either AMD or Intel chips will suffice in terms of quality. Both make a very good, and stable CPU.

Same here and I've used Hyperthreading while overclocking as well. Adobe Photo Suit, Roxio and or Nero apps as well. Something many reviewers will not event test. I've ran Creative's WAV studio, Mix Master, Photo Shop and Windows Movie Maker at the same time as well without problems. Without HT I did these one at a time and but multi-tasking was slower, almost like an AthlonXP.

Quality parts are very important for any DIYer. But even new untested Cards and Drives seem to just have an easier time working on Intel Chipset based boards.

Donnie27

 

[Josh_B]

I never had any problems with X-Fi, a few of my nForce4 friends had to wait on BIOS updates. I've just had fewer compat problems with all kinds of cheap or good stuff.



Same here and I've used Hyperthreading while overclocking as well. Adobe Photo Suit, Roxio and or Nero apps as well. Something many reviewers will not event test. I've ran Creative's WAV studio, Mix Master, Photo Shop and Windows Movie Maker at the same time as well without problems. Without HT I did these one at a time and but multi-tasking was slower, almost like an AthlonXP.

Quality parts are very important for any DIYer. But even new untested Cards and Drives seem to just have an easier time working on Intel Chipset based boards.

Donnie27

You can hardly fault AMD for not having the critical mass necessary for testers to use a variety of chipsets designed for A64's. Creative, like most other companies, appears to spend much more time working with Intel chipsets with the assumption that 90+% of their users will use the same.

 

[M3at]

Totally can't agree on the Hyperthreading thing. On the comment about S754 and S939 AMD chips, I agree. As long as you build the computer with good components, either AMD or Intel chips will suffice in terms of quality. Both make a very good, and stable CPU.

Alot of folks swear by Hyperthreading, but I guess for what I do, it isn't neccassary. It could be that this system is a low budget Dell system..heck, I don't know why Hyperthreading would cause instability issues with this system. Realistically, I should build a quality Intel system and see if I notice more of a difference. BTW, I loved my P3 600e, it's still going strong!

 

[Dan_D]

Alot of folks swear by Hyperthreading, but I guess for what I do, it isn't neccassary. It could be that this system is a low budget Dell system..heck, I don't know why Hyperthreading would cause instability issues with this system. Realistically, I should build a quality Intel system and see if I notice more of a difference. BTW, I loved my P3 600e, it's still going strong!

I swear by HT to. I notice a huge difference with it. The X2 and SMP Opterons do the same job and even better. So I am missing nothing anymore. I have the best performance I can possibly afford right now and I am happy.

The Intel HT chips are fine, until you try and game with them. They just didn't give me the performance I wanted. Plus at the time I ditched my P4, there wasn't an SLi motherboard available for Intel.

 

[E4g1e]

The Intel HT chips are fine, until you try and game with them. They just didn't give me the performance I wanted. Plus at the time I ditched my P4, there wasn't an SLi motherboard available for Intel.

And even today, the Intel processor boards with SLI are still immature. SLI'ing two cards gives you only a ~25% increase in 3DMark05 bench scores over a single card in the same rig, whereas AMD SLI configurations typically deliver at least a 50% increase in 3DMarks over a single card.

No wonder why Intel has chosen to use rival ATi's Crossfire system over NVIDIA's SLi in their high-performance chipsets.

 

[Donnie27]

You can hardly fault AMD for not having the critical mass necessary for testers to use a variety of chipsets designed for A64's. Creative, like most other companies, appears to spend much more time working with Intel chipsets with the assumption that 90+% of their users will use the same.

True!

You'd think nVidia would have done some testing. The main issue was just a BIOS flash away. That wouldn't stop me from building my AMD based system at all. I'll still buy Intel when Intel fits that need. If Intel really supported Dothan/Pentium-M, my next Gaming rig would have it instead of a 3500 and a 7800GT.

Donnie27

 

[Donnie27]

And even today, the Intel processor boards with SLI are still immature. SLI'ing two cards gives you only a ~25% increase in 3DMark05 bench scores over a single card in the same rig, whereas AMD SLI configurations typically deliver at least a 50% increase in 3DMarks over a single card.

No wonder why Intel has chosen to use rival ATi's Crossfire system over NVIDIA's SLi in their high-performance chipsets.

Maybe it has something to do with nVidia blocking or breaking support in its NF4IE drivers? Eventhough Intel + SLI is still slower than AMD + SLI, it is still way fast enough that bragging about WHO's faster is almost moot. So if Intel 's i975X supports ATI and nVidia, that puts nVidia at a disadvantage. It gives Intel users yet another reason to "go with Intel". Then they could have the freedom to do SLI with ATI or nVidia's Video Cards.

Tom's Preview

Even VIA has a working SLI config

http://techreport.com/reviews/2005q4/via-k8t900/index.x?pg=5

We all know VIA will ship a better version of the IE (Intel Edition) than the Buggy nVidia boards. With i975X and K8/I?/T900's, nVidia's boards will seem limited.

Donnie27

 

[ken328]

I'd like to add something on this topic. I just replaced my AMD X2 4200+ machine with a P4 830 one. Many ppls have a concept of "AMD is good on gaming", it is not really correct. FPS is just a part of the total gaming experience, stable, error free also parts of it. I used almost half year to "fight" with my X2, wasted alot money on it, but finally, it just not working fine, even with those extra trouble patches: AMD proc. driver, MS 896256 hotfix + regedit. It is the problem with the CPU itself, I tried 3 boards with 3 different brand chipset plus over 10 differnt versions of bios, still no luck to fix all the problem appear in gaming. Only thing that really worked for me was to set the affinity of my game's thread to run in only 1 core.

I hate the extra trouble, setting when I need run a game everytime, so I decided to give up the X2 finally and built an Intel 830, no need to say, all the gaming problems appear on X2 magically gone. Now, maybe I lost 30% of peak FPS on my games, but at least I can save alot of time and brain power to make them running smoothly.

 

[Dan_D]

And even today, the Intel processor boards with SLI are still immature. SLI'ing two cards gives you only a ~25% increase in 3DMark05 bench scores over a single card in the same rig, whereas AMD SLI configurations typically deliver at least a 50% increase in 3DMarks over a single card.

No wonder why Intel has chosen to use rival ATi's Crossfire system over NVIDIA's SLi in their high-performance chipsets.

Have you ever done extensive testing with an nForce 4 SLi Intel Edition chipset based board? Well I have. I had a board based on the nForce 4 Intel Edition chipset for about three weeks. It was stable and performance wise, was comparable to Intel chipsets. I wouldn't call it immature. Sure it hasn't had the market penetration of the nForce 4 AMD chipsets. The Intel version isn't as old either. Benchmarks show that Intel based SLi systems show less performance gain, and generally perform like a machine equipped with a A64 3000+ or 3200+ in games. This isn't the chipsets fault. I think this is due to the difference in how Intel's perform in games. They simply aren't nearly as fast. So as a result, the Intel systems perform worse.

The motherboard I tested was very stable, and ran games fine. The 3D Mark scores weren't impressive by any stretch, but they weren't horrid either. I certainly don't remember SLi not scaling as well with it. I just remember overall gaming performance being worse. As is typical with Intel processor based machines.

Intel didn't choose Crossfire at all. Intel was going for Crossfire and SLi certification. nVidia chose not to certify Intel's chipsets, and dropped support for them. Probably because they wanted to increase market demand for their own Intel Edition chipsets. It was a business decision nVidia made. Nothing more.

 

[gerbiaNem]

They should just agree on one standard, this is becoming a nightmare for consumers. Some noobie buys a motherboard with 2 SLI slots to find one of his cards just doesn't work...