OC-ing compromise...

[m25]

To put it simple; the only way to get 2.6GHz (or more) out of my 4200+ is to use an IDE HDD and forget about the SATA ports because they all lock up after HTT 225. Now, I want to do it but the only question I have is; does going to HTT 237 (for 2.6GHz) have any chance of damaging the IDE HDD I'll be using?!

 

[prong]

To put it simple; the only way to get 2.6GHz (or more) out of my 4200+ is to use an IDE HDD and forget about the SATA ports because they all lock up after HTT 225. Now, I want to do it but the only question I have is; does going to HTT 237 (for 2.6GHz) have any chance of damaging the IDE HDD I'll be using?!

Get a 4800 and use your SATA.

 

[m25]

Well, for all the rest, the chipset is perfectly stable and I have heard that some SATA ports lock automatically after a certain amount of OC-ing to prevent data corruption.

 

[m25]

To put it simple; the only way to get 2.6GHz (or more) out of my 4200+ is to use an IDE HDD and forget about the SATA ports because they all lock up after HTT 225. Now, I want to do it but the only question I have is; does going to HTT 237 (for 2.6GHz) have any chance of damaging the IDE HDD I'll be using?!

Get a 4800 and use your SATA.
:roll: ...yes, and spend like $200 for it, or get an FX-60....no way man!

 

[Dahak]

Use the memory devider to lower your ram speed and drop your hyper transport multiplier from 5X to 4x.That should fix your oc'ing problem.Goodluck.And try increasesing your cpu voltage.Just a little at a time mind you.I've found that even a slight increase in cpu voltage can make all the difference between being stable or not.

Dahak

AMD X2-4400+@2.6 TOLEDO
EVGA NF4 SLI MB
2X EVGA 7950GT KO IN SLI
4X 512MB CRUCIAL BALLISTIX DDR500
WD300GIG HD/SAMSUNG 250GIG HD
ACER 22IN WIDESCREEN LCD 1600X1200
THERMALTAKE TOUGHPOWER 850WATT PSU
COOLERMASTER MINI R120
3DMARK05 13,471

 

[1Tanker]

To put it simple; the only way to get 2.6GHz (or more) out of my 4200+ is to use an IDE HDD and forget about the SATA ports because they all lock up after HTT 225. Now, I want to do it but the only question I have is; does going to HTT 237 (for 2.6GHz) have any chance of damaging the IDE HDD I'll be using?!

Get a 4800 and use your SATA.
:roll: ...yes, and spend like $200 for it, or get an FX-60....no way man!Get an E4300. :tongue: j/k.

 

[Flying-Q]

Use the memory devider to lower your ram speed and drop your hyper transport multiplier from 5X to 4x.That should fix your oc'ing problem.Goodluck.And try increasesing your cpu voltage.Just a little at a time mind you.I've found that even a slight increase in cpu voltage can make all the difference between being stable or not.

Dahak

Agree.

Remember HTT speed is not supposed to go over 1000 so adjust HTT value and multiplier to give <=1000.

Q

 

[dr_kuli]

Man, lower your multi on HTT to 3x! 4x225=900, this board can do 800 ONLY! Second thing is PCI lock, which seems not to work. Set PCI to 33 MHz directly, PCI-x to 100-101, AGP to 66-67, not any Auto. If it doesn't help - use any SATA controller on separate PCI card, sometimes SB integrated SATA ports are not locked (e.g. Nforce3), when PCI slots ARE!

And remember, CPU has NOTHING to do with your problem.

 

[m25]

Use the memory devider to lower your ram speed and drop your hyper transport multiplier from 5X to 4x.That should fix your oc'ing problem.Goodluck.And try increasesing your cpu voltage.Just a little at a time mind you.I've found that even a slight increase in cpu voltage can make all the difference between being stable or not.

Dahak

I have NO problem with the RAM divider or HTT multi; I have my RAM running @ an overall 398MHz with dividers + OC and the HTT multi is already @ 4x (my chipset is an old one and won't allow for more anyway). The problem is that after HTT225, the SATA HDD becomes unbootable; I plug in an old IDE one and I can do all the way up to 2.6GHz, stock Vcore, stock cooler but I want to know the risks of data corruption and HDD failure at these levels.

 

[dr_kuli]

PCI lock seems not to work. Set PCI to 33 MHz directly, PCI-x to 100-101, AGP to 66-67, not any Auto. If it doesn't help - use any SATA controller on separate PCI card, sometimes SB integrated SATA ports are not locked (e.g. Nforce3), when PCI slots ARE! Once again, when your FSB is 225, your HTT is 900 (4x multi) which also may lead to data corruption, because this chipset cannot do more than 800 in specs.

I've seen a lot of similar problems on Nforce3 chipset boards. PCI lock doesn't work on two SATA ports, integrated in Southbridge, but DO on another two ports from separate SATA controller integrated onboard. Also, the lock DO work on ANY PCI card in ANY PCI slot! Your SATA ports are ALL integrated in Southbridge and seem to be UNLOCKED.

 

[dr_kuli]

Just to make it clear - you have the perfect symptoms of frequency unlocked ports on some or all PCI busses on your mobo. If you set FSB at 225 and your SATA ports are unlocked, you will have 37.5 MHz on them (instead of 33.33), which is their limit. I can't guess which other ports are affected, but it might be, that all of them. You cannot damage ANY HDD with it, but data corruption will occur for sure. IDE (PATA, not SATA) ports are not so sensitive, sometimes they work even on 41.5 MHz on PCI bus. But instead of checking when data corruption will occur, try to LOCK PCI busses at 33.33, and PCIe at 100 (AGP at 67). That's all.

 

[m25]

I have locked the AGP/PCI to 66/33MHz and locked the PCIE bus but I still have a lockup after HTT225; doesn't help If I drop the multi to 3X or run my RAM @ 166MHz. I bought a SATA PCI adapter but 1-it has a stupidly long shape and runs into a chipset cooler I have added myself 2-even when I removed it, that thing is not supported by XP x64.

 

[dr_kuli]

Hmmm, most of them are supported in XP64 by drivers from chip manufacturer, e.g. SiliconImage, JMicron. Did you check it?

 

[dr_kuli]

I'm not sure at which of my statements you replied??? I try to explain the same as you - there is no PCI lock on this board, in my opinion.

 

[rodney_ws]

Well, for all the rest, the chipset is perfectly stable and I have heard that some SATA ports lock automatically after a certain amount of OC-ing to prevent data corruption.

I know this sounds like voodoo magic, but I was having the same problem with my X2 4400+. I was told to try SATA ports 3 and 4 and I did make it to 2.6... now I do need to be clear that I was changing a LOT of things at once trying to get to 2.6 and didn't really bother going in an orderly manner. However, to this day... my WD Raptor is on SATA port 3. It'll take you 2 seconds to test this.

 

[dr_kuli]

Was it the same mobo/chipset?

 

[rodney_ws]

Mine is an Asus A8N-SLI bought ~ June 2005.

 

[m25]

Well, for all the rest, the chipset is perfectly stable and I have heard that some SATA ports lock automatically after a certain amount of OC-ing to prevent data corruption.

I know this sounds like voodoo magic, but I was having the same problem with my X2 4400+. I was told to try SATA ports 3 and 4 and I did make it to 2.6... now I do need to be clear that I was changing a LOT of things at once trying to get to 2.6 and didn't really bother going in an orderly manner. However, to this day... my WD Raptor is on SATA port 3. It'll take you 2 seconds to test this.
Have tried this too; same problem with all the ports... I know that what NovaThunder says is right about data corruption; I don't want my HDD to scream like a b!tch or to corrupt the data I've worked on for a whole month, that's why I asked about it.
OK, I won't die for 200MHz and will feed my CPU-sick brain dreaming of my next (at least quad) system till I get it ....
:idea: well... what If I set the IDE HDD to work in UDMA5 or UDMA4 mode; something like 100MB/s or 66MB/s...It could act like a memory divider, or not?!

 

[dr_kuli]

@ NovaThunder

With all the respect - what have NB to do with it??? There's nothing like NB in this chipset, only IGP. Even if it would be - HDDs (SATA or PATA doesn't matter) are connected to SB, NOT to non-existent NB. Check http://www.via.com.tw/en/products/chipsets/k8-series/k8m800/ or any other A64 chipset.

@ m25

It could NOT act like a memory divider, artificially degrading HDDs to ATA 100/66 will not change their REAL transfer, only theoretical maximum of interface speed.