Are Athlon sytems really more unstable than P4/P3?

[inqztive]

Okay let me first mention to you that I do not want another fight over this arguement. There have been enough of that. I want only technical posts on this. I personally like Athlon for the price/performance ratio. I have build more than 20 Athlon/Duron systems compared to only 1 Pentium. So I don't have enough statistical data to comment on pentium system and will restirct my opinion only to Athlons. However people who have enough experience on that is very welcome to post your opinion.

Now the most common problem I am having with quite a few Athlon systems is they reboot on their own if left idle for sometime. i.e allowed to hybernate. when I deactivate "auto restart in case of problem", it doesn't come out of Hybernating probably due to failure of the video card initialization ( the monitor says, "check the cable" , the same message you get when you unplug the video cable). These system has a few things in common and I am trying to figure out which one is the culprit.

1) All Windows 2000 system
2) All Athlon systems
3) All VIA chpset systems. (either KT 133 or KT133A, mobo manufacturers vary, but mostly ECS K7VZA and Iwill kk266)
4)All ATI video card (Expert 2000 or Radeon LE)

Latest VIA drivers and patches and SP 2 for windows 2000 are already installed. Looking forward for your opinion/solution and it would really help me to satisfy a few unhappy customer.

 

[Shack70]

system resets like that are sometimes because of bad/underpowered power supplies.....you may want to look into that

 

[inqztive]

One of the possiblities. But the system resets only when it left idle and hybernating and consuming the least amount power. Moreover I use 300W power supplies. So underpowered power supply is a very improbable cause. I am wondering is anyone else has similar problem with Athlon/Win2k systems. You leave on overnight and find that it has restarted when you check up in the morning.

 

[Ausm]

Anandtech Servers use AMD components so they can't be that unstable

Ausm

 

[Ryan]

I think the majority of the problems are due to Via chipsets. I would be bad to base an opinion on which has the most problems by looking at anandtech. The majority of the people here have AMD systems, so that's what the majority of problems are gonna come from.

 

[Whitedog]

Generally, you DO hear about more problems with Athlon systems than Pentium systems, so my answer would be YES... in general.

HOWEVER, most of the problems you hear about are fixable through driver updates...etc... so the real answer would be AMD systems are a bigger PITA (pain in the ars) than Intel systems.... but I wouldn't say less stable, because once you get all the bugs worked out, they run just as good as anything else.

a LOT of people don't even have any problems with them at all...


I think the problem generally stems from just being the "new kid on the block"
Pentiums have been around since BEFORE Windows 95. Athlons came out Just before Windows 2000 was released... And it took the Hardware companies a LONG time before they had the balls to make platforms to support them... Operating systems have a difficult time with the AMD systems simply because they were wrote before the AMD platforms were in the masses.

Things are changing now.

When I loaded XP, I didn't have to do anything when I was finished except download a better driver for my GeForce (which goes without saying)

AMD platforms are just coming into full bloom. Don't be quick to judge based on what you've "seen in the past"

.02

 

[dullard]

Ok you are mixing two different concepts.
1) Are AMD computers more stable or less stable than Intel computers? This has been debated to death, and I won't go further into details.
2) Hibernation.

I use 4 desktop computers every day. One won't sleep at all even if I want it to. One won't ever fully wake up - the video card is the problem here. One has no difficulties at all. The final computer appears like it wakes up - but hangs everytime I connect to the internet after it sleeps.

Here is the problem: hibernation was meant for laptops and only laptops. Only laptop parts are guaranteed to come out of sleep. Non laptop videocards and other components are not guaranteed to wake up. Some may be designed to do it, but the manufacturers rarely if ever test it. If you watch forums for a long time you will see a trend of two common questions. The number one most common question: Why does my clock lose/gain time? The number two most common question: Why doesn't my computer wake up? Answer: at least one component wasn't designed to hibernate (usually it is the video card that fails to wake up). Conclusion: don't let desktop computers go to sleep - they weren't designed to do it, they weren't tested to do it. This has nothing to do with AMD vs. Intel.

 

[inqztive]

So you are saying the Video card might have problem with hybernating. And that video card irrespective of being in an AMD system or intel system will have problem with hybernating or coming out of it. I don't have too many experiences with Pentium systems. So I was curious to know if this problem is a "Athlon problem" since i have seen this on quite a few systems. But what you say makes sense to me as it looks like it's a video card problem as the message on the monitor being check the cable for connection (as if it is being unplugged from the computer). But if that is the problem can this be fixed with updated driver (although I believe I have the latest drivers) or moving to windows XP instead of just stop hybernating?

 

[flyerI]

It seems to me that alot of us who like to build our own rigs like AMD because of the perfomance not to mention the price. Some of us like to "game" big time and almost all of us like to tweak every last mhz, fps, etc. out of our system. We are always looking for the next piece of hardware BIOS setting, or software that will make a improvement only noticeable in a benchmark test. This in itself is going to lend itself to some instability problems. I really do not see nearly as many posts in this or other forums about Intel systems. Does that mean that they are more stable and no body has problems? I don't think so. I think the fact that this is a forum for computer enthusiasts, and AMD is largely the choice for enthusiasts, we will not see as many problems with Intel rigs here. I have a Duron and an Athlon that are rock solid. If they do something weird it is because I caused it. I don't OC and I run conservative mem cas settings and generally don't try to get every last thing out of them. I also have an Intel machine that I have tweaked until it became so unstable it was of no use at all. I do agree with the previous post that desktop rigs do not generally like to sleep. I have one that will not and one that will. I have seen this behavior in both AMD and Intel so I really can't give you advice on how to fix you problem except to set it to never standby or hybernate.

 

[piasabird]

I think Intel processors are a lot more stable than anything with an AMD processor. The reasons may surprise you. INTEL puts a lot more research into developing its own chipsets and testing equipment. For this reason they are sometimes slow to release their CPU's.

Example I purchased a FIC AD11 and an AMD 1.2 Gig processor, and tried to get it to work with an ATI Radeon VE 32 MB DDR video card. I have gotten lots of lousy results, slow speeds and crashes.

Counterpoint: I purchased an ASUS TUSL2-C (PIII) with a Celeron 1.2 Gig Processor with the 256kb L2 cache and the 0.13 micron die process. I put an extasy (OEM ) (video out only) 32 MB DDR video card on it and I was surprised how well it ran. I had never to this point ever purchased a Celeron. This card runs with my old PC100 128 MB of memory just fine. It is designed to handle the PC133 memory, but I put the old memory on it. The board uses the intel Chipset for PC133 Ram. This is an excellent package. I purchased the moterboard and boxed CPU from www.adamant.com/ and I got the rest from www.newegg.com. I used an old case. The video card cost me $49.00 for a Geforce 2 GTS V Chip. I put in a game from SSI called The Moon Project and it ran like a bat out of hell. I also put a nice Enermax adjustable fan that can spin at 1000 - 3000 rpm, and a $18.00 Creative (OEM) sound card. You can turn the fan down if it makes too much noise.

I see no downside from this cutting edge ASUS PIII board with the INTEL chipset and Celeron Processor. I would recommend it to anyone. Asus had an excellent setup disk. I changed nothing in the BIOS and installed a patch they used to make Win98 see the hardware on the motherboard that came with the motherboard. I never have changed but just a couple settings for the AGP port.

 

[MadRat]

Hibernate probably drops the voltage to parts throughout the system. The resumption probably doesn't wait until voltage is back to full flow.

 

[MrGrim]

Anything that has to do with VIA sux. I'm talking from experiense.

 

[Priit]

Stability differences between AMD and Intel systems are mostly due to software issues ("cought" M$ "cought") or misconfigured hardware (too weak PSU's etc). I don't have had ANY stability problems with AMD/VIA systems over the years. My friend works in place where they have over 500 duron/athlons machines and he has no complains about stability either. Sorry that I can't help about Windows hibernate problems, I haven't used W2k/XP much.

 

[x86]

I must say that AMD doesn't spend that much money on R&D which is the primary cause that AMD is incompatable, unstable, and sometimes unreliable. This is one of the major factors that has kept AMD out of the Server market. Conversely, Intel spends about $7,000,000,000.00 on R&D, and makes sure that all of it's processors are compatable, and safe to work with. This is one of the factors why Intel chips may cost a bit more than AMD's, but it isn't too much to ask for when you spend a majority of your time on the computer and don't want to deal with non-sensical problems.

-x86

 

[MadRat]

I sense a small waft of troll in that last post.

 

[jh0sken612]

<< Anything that has to do with VIA sux. I'm talking from experiense. >>



How is this so? I owned/built a few different PCs that use VIA chipsets, and I haven't had any problems at all.

ABit KT7-E, etc. etc.

Thats funny... who makes better chipsets... Intel? (Laughs)

 

[AGodspeed]

I must say that AMD doesn't spend that much money on R&D which is the primary cause that AMD is incompatable, unstable, and sometimes unreliable.

AMD processors themselves aren't believed to be the cause of general problems, the chipset and OS/software are usually the culprits...

 

[sandorski]

no

 

[mrman3k]

Let me give you some of my more recent data on building machines as my part time job.

18 Athlon/XP/Duron Systems
11 Pentium III/P4
2 P4 Northwoods

All of the systems have been rock stable, the only one that is not very stable is one that is an Athlon XP but has almost every component overclocked (the guy that ordered it was nuts, he put certain speeds he wanted attained even if it meant no stability, as long as it could run the benchmarks he was okay with it.) I would say that the majority of Athlon systems that I get that are crash happy are the ones that are built by amateurs that do not use enough power, or cooling for their processors. However I have seen this a few times with Intel processors, once this person had a Celeron that actuall BONDED to the heatsink. When I pryed the heatsink off half of the core was stuck to the heatsink. The othertime was P3 that looked like it had melted but not like the melting in the Tom's video, but basically it was no thermal paste which caused it to melt. Plus that guy had the cherry on the whip cream applied when he did not have the computer hooked up to a surge protector and well, there was a surge which fried it.

So I would say both systems are just as stable as long as you pick the highest quality and well recommeded parts.

 

[glenn1]

<< AMD processors themselves aren't believed to be the cause of general problems, the chipset and OS/software are usually the culprits... >>



Add overly-aggressive overclocking techniques onto that list, and i'd say you've covered 99.9% of the supposed "AMD instability problems." I'd say that problems with the CPU or its architecture are so far down the list of things to cause problems, as to be essentially zero. Even when they are (the infamous original Pentium floating point bug being one example), the problems are always hugely and widely publicized.

 

[AA0]

I've used so many computers, and heard of so many complaints of people using desktops (intels mostly) that lock up and they lose everything on them. Its always do to windows sleep features, I disable it for them, and they think I'm a god.

Now... assume these people are computer idiots and I love intel, I'd say its a AMD chip inside that screws everything up. There is the source of your rumors...

In winxp is looks like ms finally stopped putting these features on by default, that god. I still vow that if you buy good hardware, you will not have a problem with your system. So many people assume just because the AMD chips are cheap, every part that goes into their system can be cheap too. Well it doesn't happen on intel systems that way, thats why they are stable.

 

[Tanked]

Any stability problems (or slow, buggy performance) is virtually always going to be the motherboard or some other system component, not the processor. I have to say I have a sour taste from VIA, as they're chips have given me problems all over. If a Radeon is having problems with an Athlon is going to be a problem with the motherboard, not the actual chip. In my eye, AMD chipsets are far superior, as long as they aren't matched with a VIA southbridge (which is almost always).

 

[oldfart]

<< << Anything that has to do with VIA sux. I'm talking from experiense. >>

How is this so? I owned/built a few different PCs that use VIA chipsets, and I haven't had any problems at all.

ABit KT7-E, etc. etc.

Thats funny... who makes better chipsets... Intel? (Laughs) >>


Yes, Intel AMD and SiS all make better chipsets than VIA.

 

[DaemonJ]

Some systems aren't able to sleep when:

1. Media Player 7 is installed and you install the Adaptec compact disc recordable (CD-R) plug-in. You have to upgrade your BIOS or not install the Adaptec plug-in to fix that problem.

2. If you install PGP Desktop Security 6.5.3 or PGP Freeware and you turn on the PGPDisk feature. An update is out to fix this.

3. After you install and then subsequently uninstall McAfee AntiVirus version 3.1.4. McAfee AntiVirus 3.1.4 with NetShield version 3.0.2 services is not compatible with Windows 2000. When you uninstall McAfee AntiVirus, the uninstall routine does not completely remove all of its dependent services and related files. These services are left running, which prevents Windows 2000 from hibernating.

These are just the top items. Hibernation has always been a problem for Windows. This has nothing to do with AMD or the motherboards' chipset.

 

[meson2000]

All I have to add is my own personal experience.
I have built 3 intel systems and 2 AMD systems by my own hands.
Of the 3 intel systems that I have built 2 had to have components replaced from the beginning. My first hand built Intel system
was an Intel 80386 33mhz system (Upon testing by the store I bought it from, the motherboard was dead. They gave me a new one) and my 3rd Intel system was an Intel Celeron 433 (Upon testing by the store I bought it from, the CPU was dead. They gave me a new one). Of the 2 AMD systems I have built one was an FIC KT133 board with an AMD Duron 950 and the second was an AMD Athlon 1.33mhz with an ASUS KT133A board. Both AMD systems ran rock solid from the beginning without any crashes or incompatabilties from the start. I almost gave up building my own systems until I switched to AMD and realized how easy it could be. As far
as my experience goes, AMD is more stable than Intel. No flames, could just be bad luck, but this is my own personal experience.