Rebuilt CPU is VERY slow

[rss00]

I just bought a new ECS L4VXA2 mobo, 2.5 GHz Celeron, 256M PC2100 mem, 60 G Western Dig disk, Verbatum CD R/W.

Installed Win 98 and system runs slower than the 133 Mhz PI it repleaced. No errors show up and it seems to work but is VERY slow. FSB jumper is set to 400.

I replaced the mobo but it did not fix the problem (I thought it was probably the clock running at a slow rate). Can RAM cause this? How about the processor? Diagnostic software does not indicate any hardware problems.

Probably some simple solution that I just don't know about.

Thanks for your help.

[Force Flow]

Welcome to PCMech, rss00!

Well, the PC2100 is a little slow, but that shouldn't be the problem.

What's the speed on the HDD, 5400rpm or 7200?

It may be that win98 isn't taking advantage of the hardware. You may want to consider win2000 or XP.

[Panama Red]

Here's a crazy thought. Is your cpu fan plugged in and running? Almost sounds like the CPU is running hot and throttling back to compensate.

[rss00]

Panama Red & force flow,

Thanks for the rapid response. Processor is running barely above room temp...24 degrees in a cool room. Fan is running.

Disk is 7200 rpm.

I took 3 hours to load MS office...only about 1 hour on my old P 1.

CPU runs so cool I know it is not clocking properly...not even warm to the touch.

[Floppyman]

Did you install the chipset drivers that came with the motherboard? Have you checked whether there is a bios update that might address this issue? HTH

[HAL9000]

You know.. you many need to reload the system. I recently upgraded my bench machine from a 1.3Ghz Celeron to a 2.4Ghz Celeron. It would initially run fine.. then start lagging REALLY bad. I could tell when somebody was paging me on chat as my mouse would stop for about two seconds, then the chat window would open, then the message would display, then the sound played. I had to reboot 3 or 4 times during a 7 hour work day. I tried a repair install, no effect... I finally upgraded to WinXP and all is well.

[escher]

maybe the cpu is what's going out. at some of my jobsites there are many celeron machines....who knows if this is your issue, but for some reason with those things, when they start to go out they get flaky or really slow. we replace the cpu and all is good again. <---i really dont like celerons for this reason. usually cpu's are just live or die...

[glc]

There's a reason ECS boards are so cheap. There's also a reason I won't touch Via chipsets for Intel processors. Windows 98 also has issues with processors faster than 2.1 GHz.

[Panama Red]

Originally posted by glc There's a reason ECS boards are so cheap. There's also a reason I won't touch Via chipsets for Intel processors. Windows 98 also has issues with processors faster than 2.1 GHz.

I'm curious about the issues, g. I've had 98se running on my P4PE/L board with a 2.4b and 1Gb of ram since May with no trouble whatsoever. It's set up as a dual boot, but it is on a separate hard drive. I don't use it all the time but when I do it flys. Connects to the web just fine through my home network with broadband and maintains the same 3100+Kb/s download speed.

[melloman]

I had a Shuttle board and a 2 ghz Celeron, and the via chipset board sucked horribly. I finally gave up when it would act funky just booting up. I put the cpu into an ASUS board, and it ran like a top, and ran at 2.66 Ghz, and was more stable than the shuttle at 2 gig.

I have given up VIA as a chipset maker, SiS seems to do okay on cheap chipsets. For fast machines, I like Intel chipsets.

[yellohut]

Originally Posted by: Panama Red I've had 98se running on my P4PE/L board with a 2.4b and 1Gb of ram since May with no trouble whatsoever.

Windows 98's timings aren't compatable with the faster processors.. I'm not sure if a patch will fix this..
It's the same as win95.. it freaks out if you put a cpu faster than about 1.2 ghz.
Maybe you got lucky..
Also, Win 98 can only use 512 MB of RAM, it may recognize it all, but only addresses the first 512.

As for the slow running Cely, Make sure all your drivers are installed for the mobo, and make sure the bios is up to date..
also, have you installed all the patches and upates for windows 98?

[melloman]

You might have better luck with win2k.

[escher]

yeah no kidding. why would you use win95/98 with a machine of that power anyway?

[melloman]

Let me guess.... cause 95 won't boot on a P4?

Just kidding. I have used 98 for a long time, and I moved to mostly 2000 because it works, is steady, and doesn't have the quirks of XP.

Although, I have Red Hat 9 running nicely on a celeron 800 with almost no effort at all, and Samba is a lot better for connecting to windows shares than it used to be.

[Panama Red]

Think the reference to problems over 2.1 (not 1.2 as stated by yellowhut) comes from problems that were corrected by a MS patch for the NDIS.VXD.

[rss00]

It seems like everyone if forgetting that the system is SLOW but working without "flakey" Win98 problems. The processor temp is about 25 - 28...not even warm to the touch. Wouldn't this indicate clock speed? Can windows slow the clock down?

Spent all day loading new mobo drivers, BIOS, etc and now it doesn't acknowledge the keyboard and mouse clicks after boot is complete. Mouse pointer works but that's all.

Progress is great if going backwards is the right direction.

[HAL9000]

Originally posted by escher maybe the cpu is what's going out. at some of my jobsites there are many celeron machines....who knows if this is your issue, but for some reason with those things, when they start to go out they get flaky or really slow. we replace the cpu and all is good again. <---i really dont like celerons for this reason. usually cpu's are just live or die...

You would have to prove that to me... having been thru thousands of processors... and only ever having 5 bad ones outta that....I don't see how you could just be having Celerons "die"

I'm still thinking your best bet is back everything up before its too much of a mess.. format and reload.

[escher]

celerons/p3's will do this. not sure about the newer ones, but with these it definitely is not a "live or die" thing. they will run like crap for ages before they finally kick the can. the Slot 1 cpu's seem to be the worst about this. perhaps its something to do with the off-die cache or the PCB.

[HAL9000]

Again.. you would have to prove that to me...I literally have to call bogus on a statement like that... like I have said before.. personally built in excess of 5000 machines... not including upgrades or just servicing... you're getting all these "flaked out" PIII's and Celerons and I have had maybe 5 bad CPU's EVER, not all of which were Intel either.

[rss00]

Originally posted by Panama Red Think the reference to problems over 2.1 (not 1.2 as stated by yellowhut) comes from problems that were corrected by a MS patch for the NDIS.VXD.

You're the man!!!

Spent 4.5 hours reloading windows to recover from yesterday's problems encountered when reloading drivers etc.

After reloading, went into the BIOS and enabled L1/L2. Which I originally disabled. UGH Disabling this is the only way it would install windows. Thought there was only one cache in celeron. DUH

Anyway, with L1/L2 enabled, booting windows caused "windows protection error" while loading ndis.vxd. Booted in safe mode and performance was great. Booted again using line by line prompts and deselected ndis devices as they were being enabled. System booted fine and performance is great.

Found exact problem described on MS web site (as panama red alludes to) for processor faster than 2.1G but it appears to require a call into microsoft to get a fix. They acknowledge it is a problem with the OS but still require a service call rather than just posting a download. Unless anyone else has seen this and has the patch, guess I'll have to wait until monday and give them a call.

Thanks for all the help.

Microsoft's summary is:

SYMPTOMS
When you are installing Windows 95 or Windows 98 on a computer that has a CPU that runs at 2.2 gigahertz (GHz) or faster, you may receive the following error message:

While initializing device NDIS: Windows protection error
CAUSE
The timing calibration code in the Network Driver Interface Specification (NDIS) driver causes a divide by zero if the CPU runs at 2.2 GHz or faster. This problem does not occur with CPUs that run at 2.1 GHz or slower.
RESOLUTION
Windows 98
A supported fix is now available from Microsoft, but it is only intended to correct the problem that is described in this article. Apply it only to computers that are experiencing this specific problem.

[Panama Red]

It appears this problem/hot fix apply to Win 95 and 98 First Edition. Are you running the original 98, First Edition? If so, you may want to find a copy of 98 Second Edition. That's what I'm using and the problem doesn't exist for me. Just helped install 98se today on a new system using a 2.2 Celeron and after all the Windows Updates it purrrs!

[yellohut]

Originally posted by Panama Red Think the reference to problems over 2.1 (not 1.2 as stated by yellowhut) comes from problems that were corrected by a MS patch for the NDIS.VXD.

my bad.. must've been a typo..

glad to hear you've got it all worked out..
and i was unaware about the patch, thanks panama.

(btw, it's yellohut, no "w" )

[rss00]

Yup, old version of 98 I got with the P 133...let me see...100 years ago. If I can't find 98SE for free, will probably rob a bank for $90 and buy XP.

[escher]

how am i gonna prove that to you? this is called a forum. if you want proof go research it.

[HAL9000]

I have researched it and back it with several years of experience with usage of thousands of processors... to say that Celeron's "just die" or that PIII's "just go flakey" is a total joke.... If it were really the case, shouldn't I have run into a few of them? Since I am Intel biased, it would be just as ridiculous for me to say, AMD chips are responsible for most house fires... they just do that.

[bewitt7]

How is your HD hooked is it hooked at master or single. Is it on a hook up alone If so it should be single.

[Panama Red]

Originally posted by escher how am i gonna prove that to you? this is called a forum. if you want proof go research it.

Proof of anything comes from reference to previously documented tests or statistical surveys. Simply stating you believe something to "be so" because you heard it from someone else or read it in another forum does not validate a statement. Normally, when some one asks you the source of your info, you provide that source. Responding as you have is merely a smug, smart alec statement to remove responsibility of proof from you. Your credibility on this forum has been severly damaged, in my opinion. You either need to provide valid proof or we all can assume it's just hearsay.

[glc]

http://download.microsoft.com/downlo.../243199US8.EXE

There's the 98SE NDIS patch.

Hindsight is wonderful - you should have told us you disabled the cache - this will bring any processor to a near-standstill!

[escher]

Originally posted by HAL9000 I have researched it and back it with several years of experience with usage of thousands of processors... to say that Celeron's "just die" or that PIII's "just go flakey" is a total joke.... If it were really the case, shouldn't I have run into a few of them? Since I am Intel biased, it would be just as ridiculous for me to say, AMD chips are responsible for most house fires... they just do that.

yeah ok you're right *sarcasm*