Jump to content

another dual athlon system


Recommended Posts

Cesar,

 

All boxes have defects. Its whether you choose to fix them that determines if the machine will be a workstation, or somebodies desktop.

 

Run through all the tests and suggestions I gave you and the problem will be solved.

Link to comment
Share on other sites

  • 2 weeks later...

Asus motherboards aren't really that good, try csi or is it msi one of the two.

 

try to push upto to 2 mg ddr ram ( it think it has to be registered for amd dual)

 

also try the ti4600 geforce 4 128mg card. it's marginally more expensive but better.

 

really i think this is the highest spec you will get at this point in time unless you spend £1000 plus on your graphics cards and go all pro on me!

Link to comment
Share on other sites

  • 2 weeks later...

Greg,

 

How do the HCC-001'c compare to the efficiency of the SK-6's? I have been thinking of replcaing my SK-6's with HCC-001 but still using the 60mm DELTA fans. Don;t worry I;ve gotten used to the 'ducted fan noise' on my Dual AthlonMP 1600+s.

Who knows I might get those newer 2100/2200's but not sure if they will run on my TyanMP (S2460). Also is it true that the MP's will surive while the MPX will be phased out when the Opteron's come on line? Just wondering.. :)

Link to comment
Share on other sites

Just got in the system -

 

Dual Athlon MP 2000

(2) Thermaltake Volcano 7+ Coolers

Asus A7M266-D

Western Digital Caviar WD1200JB

1.5GB Crucial DDRAM ECC REG

CDRW - Plextor 40x12x40

DVDROM – Pioneer 106s 16x/40x slot load

Floppy, Zip, IntelliMouse, Keyboard, Network Card, Firewire Card

LianLi PC62

Enermax Whisper 430W (EG465P-VEFC)

Nvidia/ PNY Quadro4 750XGL

 

The system ran fine for a day and then crashed and was unable to be rebooted. Upon bringing the system in, they initially told me that AMD’s are just not stable and difficult to deal with, that this is why they always push Intel and that the cpu’s seemed to be too hot. With Asus Probe the cpu temps were idling after a few minutes at 51 and 54 and the mobo was at 38. They ran some tests and discovered that a memory module was bad (crucial ecc reg gone bad?) (is there a way to confirm whether they gave me crucial/ I only saw ecc reg on the sticks?) – they then decided to try the standard heatsink and fan that came w/ the cpu’s. With both of these changes, they say the system is running fine. I have yet to pick up the system again.

 

I have done all my research online and have read repeatedly that a high end processor (whether AMD or Intel but especially AMD) these days require special cpu coolers – that the standard cpu heatsink/fan combos supplied by AMD for instance are not satisfactory. My questions are as follows –

 

1. Is this true? Why would AMD and Intel supply a heatsink/fan to their processors that are not sufficient for cooling their respective processors?

2. Should I reattach the Volcano 7+’s? Keep the standard heatsink/fans?

3. The Volcano 7+’s are intolerably loud on High or Medium. I asked Thermaltake techs and they said Low would provide sufficient cooling – on low at 28 CFM/3000rpm do you think this is correct?

 

Thanks for any input,

 

Brian

 

PS> Thinking I should have gone with the AX-7 and 80mm fan combo for better cooling and noise performance. However, I did not want to deal with the modifications to the capacitor/ cpu socket.

Link to comment
Share on other sites

Hey guys i noticed that when booting up my system I get an '*' before CPU1 it looks something like this:

 

CPU 0 : Athlon MP 1900+

CPU 1 :*Athlon MP 1900+

 

I had never noticed it before, is it suposed to be that way ?, or is there something wrong with my baby ?! JEFF_btw, my system's temp is ~40,41 all the time and MB 30 celcius.

Link to comment
Share on other sites

CPU 0 : Athlon MP 1900+

CPU 1 :*Athlon MP 1900+

 

Running just fine there! Its supposed to do that. Its basically the XP detection phase. (At least under the new bios). Remember if you have Dual XP's and got them to work, do not, I repeat, do NOT flash your bios to a later revision..

 

Ok there's question 1, here they come. :)

Link to comment
Share on other sites

Originally posted by Arnold Gallardo:

[QB]Greg,

 

Hey Arnold,

 

The Sk6's are fantastic heatsinks. Their chieft weakness (and strength) is the density of their copper fins. Because their so tightly packed and dense, only extremely high CFM fans (delta's) can really make use of the heatsinks. If you drop the cfm down (to something like 18) the heatsink really drops off in performance.

 

The HHC-001's use a slightly broader fine arrangement, instead using the heatpipe technology to help dissipate heat to the top of the fins. I've found with the same CFM fan they're probably on the same performance level (delta vs delta) but they are far MORE efficent with slower spinning, quieter 60mm's or 80mm funnels.

 

The BEST one I've seen so far is the thermalright AX-7, which by default uses 80mm fans. Its quiet, its efficent, and currently one of the top performers on the market.

 

The 2100+'s should run on the tyan, tbred is not MP, so probably around 2200+ or 2300+ We'll start seeing some weird stuff happening. Should still work with our boards though, until the FSB switch.

 

MPX will not be phased out when Clawhammer/Hammer come out. Believe it or not, MPX is going to be the budget line :) .

Link to comment
Share on other sites

Brian,

 

Did you setup the system or did someone else?

 

Here are direct responses to the questions.

 

"they initially told me that AMD’s are just not stable and difficult to deal with, that this is why they always push Intel and that the cpu’s seemed to be too hot. "

 

Oem dealers don't like AMD systems, because it reveals how utterly crappy and incompetent they are. With an Intel chip, you can usually not pay attention to what your doing, and the system will run slower, and a bit unstable, but it'll still run. AMD on the otherhand will let you know immediately that someone tried to rip you off, or didn't install heatsinks properly, or used crappy components or installation procedures. 9/10 Oem's use generic slop for thermal paste and have no inkling of how to actually apply it. To them its just a rush order to get it through as quickly as possible, and could care less if the system crashes.

 

Secondly....Intel systems cost much more. This means they can charge more for intel systems, and thus make a higher profit. Their biased towards making a profit, and will continue to make recommendations based on whatever gives them the largest of said profit.

 

"With Asus Probe the cpu temps were idling after a few minutes at 51 and 54 and the mobo was at 38."

 

Those temps aren't bad, they aren't good either. Sounds like they could have been installed a bit better.

 

"They ran some tests and discovered that a memory module was bad (crucial ecc reg gone bad?) "

 

If its 3 dimms of unbuffered, there is a CHANCE it might work, but would eventually crash. Its possible that a dimm or two was bad. Always run www.mushkin.com/support memtest when you get new ram. Even if you get it from mushkin (they tell you to do thus, damage can occur during shipping)

 

"the standard heatsink and fan that came w/ the cpu’s. With both of these changes, they say the system is running fine. I have yet to pick up the system again."

 

The AMD retail box recommends against using the standard amd heatsink for anything other then testing. It actually tells you in the manual to purchase a 3rd party solution.

 

"1. Is this true? Why would AMD and Intel supply a heatsink/fan to their processors that are not sufficient for cooling their respective processors? "

 

For testing purposes. (To make sure the processor worked)

 

2. Should I reattach the Volcano 7+’s?

 

Yes. Use artic silver III thermal paste, and apply a thin layer to each cpu EXTERIOR to the case. (Aka take the motherboard out of the case). Scrap off the excess with a credit card or blockbuster video card. (The blockbuster card works wonders). There should be just enough paste to cover the surface of the cpu so that you can't read the chip ID or numbers. I usually just make sure its a very thin flat glaze. Here's a pic before I scrapped off the excess.

 

http://www.3dluvr.com/crossbow/photos/dual/paste1.jpg

 

And there is the modification I had to do to the HHC-001 to get it to fit. (Had to bend the thumb clip back)

 

http://www.3dluvr.com/crossbow/photos/dual/modification3.jpg

 

(BTW those open funnels suck ass)

 

Its very important to mount the heatsinks outside of the case, so that you can lift the board up to a light source, and check to make sure there are no air gaps between the heatsink and cpu. If there is even a hinting of one, its important to remount and try again. I usually use alcohol and acetone to remove the thermal paste. (Acetone, then some 70% alcohol...hehe I work in a biology lab :) )

 

3. The Volcano 7+’s are intolerably loud on High or Medium.

 

Their loud on medium? I'd probably push for medium, but try low and if you get instability, try medium. If it goes away that was the cause, if it doesn't its something else.

 

Remember when u get the box back, run an immediate ram test. Best to eliminate that off the bat.

 

Hope that helps.

Link to comment
Share on other sites

"they now tell me that the reason it went bad was because it overheated. "

 

Man this place really wants to rip you off. Hahahaha. That gave me a good laugh. Ram sinks/spreaders aren't necessary for DDR because of the way the heat is dissipated. Unlike a cpu which disappates its heat across a single surface, ram distributes its heat load across all the chips. Since there are usually 8 per side, thats a hella of a surface area to release heat. Most of the heatspreaders (On system memory) are there more for show then for anything else. Sure they help a tiny bit, but there is no way that your DDR is going to overheat.

 

Now what could of happened (And this is going to be part of a recommendation below). Is that the jerks sold you some PC1600 Registered DDR. PC1600 will run at PC2100, but it'll run hotter and out of spec, causing system crashes and a variety of other odd problems. (Basically ram errors)

 

If you get ram from them again and it fails the ram test (1 error is a failure), I suggest asking for a full refund, yell about the heatsinks as well, and contact a dealer like crucial, corsair, or mushkin to get your ram from seperately.

 

Thats some total BS on the volcano 7's. Ask for a refund and if they start bitching, give them thermaltake's web url and their address. They will rma the sinks.

 

http://www.thermaltake.com/

 

Direct URL to their address.

 

http://www.thermaltake.com/company.htm

 

[ June 27, 2002, 08:30 AM: Message edited by: Greg Hess ]

Link to comment
Share on other sites

To answer Greg's question, I am having a local shop put it together. Still waiting on the new 512MB RAM module to arrive to replace the defunct one - they now tell me that the reason it went bad was because it overheated. That I should buy these cheap $9 heatsinks for the RAM - I have heard of RAM heatsinks but not of this problem occuring w/ other folks. I think I will tell them no on the RAM heatsinks.

 

However, I am making the decision to go w/ the AX-7's w/ Enermax Adjustables. Will mount them w/ Artic Silver III. Will have to eat the cost on the Volcano 7+'s (they claim they cannot return it as it was a special order from wherever they ordered it from). Will run the ram test when I get it home over the weekend.

 

Thanks for all the info thusfar,

Brian

Link to comment
Share on other sites

I got the system back and it runs fine. Still need to run the RAM test. However, I went to check for a part # to confirm that I indeed did receive PC2100 ram that was Crucial/Micron, but was told by Crucial when reading them the part # that they were not Crucial modules that they were only Crucial components on a module another manufacturer had put together.

 

Does this matter? Should I return these? There is no way for me to tell whether I indeed received PC2100 RAM independent of the reseller and whoever manufactured/supplied them w/ the RAM is there?

 

Thnx,

Brian

Link to comment
Share on other sites

I'd say just run the ram test till the computer bleeds. If errors pop up, yell at people till you get the ram you actually asked for. If no errors pop up and stability is prime, I'd say just work with the system and enjoy it.

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
 Share

×
×
  • Create New...