Having had a couple of Intel-only mobos, I can say that I will never again believe the mythology about Intel. Having read what is going on with current users of the latest Intel integrated 845 graphics, I would not recommend an Intel based mobo to anyone. If you can't get it right when you totally control 100% of the mobo, you are flub-a-dub beneath consideration. Still, Intel had many good designs. Am I contradicting myself? I don't think so. You can't just buy Intel and be trouble free.
Does Intel have a better record than VIA? Yes. But there are reasons for that beyond technology.
My ASUS P3We, an Intel 810e chipset mobo, had quirky boot up problems, mostly when I did not turn it on for a few days, but once it booted it operated flawlessly - provided you used the patches game makers put out to make the 810 work with their games. I decided to give it away and so I put in a modem. Then there was nothing I could do that would keep my favorite game from crashing within two minutes. Total lock up. I spent days trying every permutation of settings, slots, drivers, and reinstallation of Win98se. Only solution: never play the game or take out the modem. With the modem, the computer also would not reboot from Windows, it would hang and lock. I couldn't let install programs do a reboot because I would just have to reinstall again. Windows would do a complete shut down though, which is a usable work-around. This modem was problem free in a VIA based mobo. So I know that
crazy, illogical symptoms are not confined to VIA.
People exist who report problems (static) with Sound Blaster and Intel chipsets. I don't know how common it is. I used to think it was a VIA-only problem when I confined myself to AMD oriented newgroups and forums. Sound Blaster has never considered it worth the effort to check their sound cards for VIA compatiblity until recently. Years ago VIA had been an upstart competitor to Creative Labs/Sound Blaster with some sound chip, if I recall, so Sound Blaster was only too pleased to screw VIA whenever the opportunity presented itself. They could hardly try that crap with goliath Intel. If there is any screwing between Intel and anybody else, Intel will do the screwing.
This site is dominated by AMD users, fans even. Somewhere you can find the statistics of those who choose to supply the info. It is something like three quarters AMD, as I recall. This is the way it goes with a lot of enthusiast's sites. True, they are stuck with VIA, sort of, rather than being fans. Are these technically knowledgable and technically competent people all fools, pathetic dupes of the VIA/AMD PR/marketing juggernaut, unable to see the reality in front of them of VIA instability? Absurd! I rather think it is Intel that is doing the successful duping. I wouldn't put it past Intel to provide free "driver development assistance" software to PCI card manufacturers that includes known confilicts for VIA without proprietary, undisclosed, trade-secret, NDA protected, non-standard aspects of Intel's PCI implimentation. It is then up to the manufacturer to fix the glitches if they can figure it out. VIA sidesteps these by supplying new 4in1 drivers as issues are added to their database. It would be reasonable to assume this is why VIA still has issues after all these years, rather than some imagined curse on VIA or some magical communion with the etherial spirits of technology that only Intel can summon.
But for every VIA user that has unresolved problems, there are 999 users that never experience one glitch. Still, 1 in 1000 can generate enough anecdotes to keep the VIA-Intel mythology spinning. Maybe there is an environment which will reveal these obscure and rare lapses that are said to differentiate VIA from Intel, but it is not my home.
Some people are noticing that Nvidia can't get Intel to license them for nForce. I have no doubt that Nvidia has a complete design ready to go for Intel, and likely had one done before the Athlon nForce. But with no license Nvidia won't enter a legal battle that is a risk which the mobo makers will not accept until it is settled. Look what VIA is experiencing with their P4 chipset. With two memory controllers, the nForce bandwidth seems a more suitable match for a P4 than a K7. That pretty much explains why Nvidia isn't going to get a license from Intel. Intel makes chipsets. You don't license yourself into a competitive disadvantage. Maybe nForce also has a legal work-alike of the proprietary PCI implementation of Intel. It kind of looks like it. If so, Nvidia could "Crush" Intel's chipset business in a year or two like what they did to 3Dfx. I hypothesize Nvidia thought it was close to an agreement with Intel: No one develops a chipset for AMD instead of Intel by preference. I further hypothesize that Intel insisted on Nvidia crippling the aspects of the chipset that made it outperform Intel's in any area whatsoever, something which SIS didn't mind doing in order to resurrect from the grave, and which for Nvidia, as well as VIA, would put sales too low to make any business sense. Another way of doing this in a contract is to require onerous royalties when the sales figures grow past a certain point. As long as you agree to be second-class, Intel has no problem licensing to you. Why would Intel agree to put itself at a disadvantage? This type of anti-competitive agreement is not illegal. For patent-holders, it is a monopoly protection afforded them by Congress. Keeping the terms of an agreement secret is not illegal: it is the usual practice; disclosing a secret agreement is illegal. We found out that even the government lawyers who prosecuted Microsoft had to observe this requirement in court.
To make this even plainer, assume Nvidia's nForce for the P3 and P4 is perfectly legal without a license and that's why Nvidia expected an agreement with Intel. Proving it in court would take time and money, and be risky. Money in legal fees which would make that row of million dollar computers at Nvidia look trivial. Time which would make nForce chipsets out-of-date by the time a ruling came down. The courts are erratic. Rulings hardly make sense in patents and rights cases. Sometimes the court will give you 9 out of 10 of your points, although the 9 would seem to imply the tenth, but you still have to settle with your competitor to do anything. It's like a card game where you've got the cards but your opponent raises the stakes so high you won't accept the risk. If you owned Nvidia, would you bet the company?
I am at this moment using my main mobo, an old VIA KT133 based mobo (ABIT KT7) with a Sound Blaster Live! I have never experience any oddities with the sound. It always sounds great whatever I do, and whatever game I try. The UT3 demo is fine. Admittedly I'm not a big game player. But when I have switched CPUs, video boards, memory, heatsinks or changed the OCing, I have run the Q2 intro for 72 hours straight without a hitch. During Holidays I have played 3D games for 16 hours straight. I torture test with Prime95 for at least 24 hours when I mess with memory timing and OCing. I have on occasion run Prime95 for three weeks straight (testing a prime). I never see any of these instabilities that people claim happen with VIA, provided I keep the CPU below a certain temp when OCing. As far as I can judge, my experience is absolutely normal for virtually everyone with a VIA style mobo. I have no idea where these freaks and weidos come from that have all these problems with VIA chipsets.
Facebook
Twitter