Hi,I'm experiencing problems during large file operations on HDs attached to either SATA1 or SATA2 running in non-RAID. The system tends to lock up while trying to copy large files, running Chkdsk on the drive while in Windows or while Defragging the drive. I have been unable to format any drive on SATA1 or SATA2 while in Windows although it is possible while in SAFE MODE (!?)My System specs:ASRock 939Dual-SATA2, AMD Athlon 64X2 3800+ (2000MHz)(K8 Manchester), 2GB Crucial PC3200 DDR400 Ram 2.5-3-3-8 (1T) (Dual Channel), (AGP 8X) Saphire ATI Radeon X800 XL (R430) 256MB (Omega 3.8.273 (Catalyst 6.7)), Hyundai ImageQuest Q17 17" TFT LCD 20ms Monitor (1280x1024x32), SB Audigy2 ZS Platinum (Drivers version 5.12.0001.1196 WHQL), WD Caviar SE16 250GB (WD2500KS) SATA2 + WD Caviar SE16 400GB (WD4000KD) SATA, ASUS DRW-1608P2S Optical Drive, Antec P150 case with NeoHE 430W PSU, CH Products USB Yoke, Pedals and Fighterstick, Plantronics Headset and Mike combo, TrackIR Pro ver. 4.1 (build 29), Logitech MX1000 Laser Mouse, Windows XP Home Edition (SP2), DirectX 9.0c, AGP Aperture = 256MBI have no problems whatsoever with the WD SATAII drive attached to the SATAII interface.I have tried an identical WD SATA2 drive on SATA1 and SATA2 but same result. I tried jumpering pins 5 and 6 to force SATA 150 but still no joy. As you can see from my specs I am now running a native SATA drive on SATA2 and it seems ok during routine Explorer operations, but as soon as I try to copy large files or defrag the drive the system locks up long before completion. I have noted also that scanning large files (>=1GB) with AVG causes the system to lockup with no escape requiring a system reset or power off. Running HD Tach Quick Bench (8 MB zones) - no probs, but the test often fails to complete without a system lockup during the Long Bench (32MB zones).This board also has an ATA133 IDE interface and compatible drives seem to work fine in all respects. Fortunately my main drive continues to work flawlessly in every respect on SATAII.By now, I am convinced that the drives are not to blame and the problem lies with the SATA interface. I have tried updating the BIOS from P1.90 to P2.30, but sadly see no change in behaviour. I do have the latest SATA drivers from ASRock installed for the board.ASRock Support have not been of much help so far. Apart from trying all of the above they can only suggest I reinstall Windows!! To me this is invariably a sign that the individual involved hasn't got a clue...but then, neither have I...LOL!I am coming to the conclusion that this board may not be stable with the SATAII port and either of the two SATA ports working concurrently. I am on the verge of reverting back to using the IDE interface as I have a 320GB Hitachi Deskstar waiting eagerly to oblige. Before I do, I thought I would post here in the hope that one of you may have encountered similar problems or recognize the symtoms and can suggest a solution.Thanks.Mike

If you're using the same cable for the tests, try a different one.You haven't said if this PC is a new build or not...i.e. have the SATA ports worked in the past?There are actually two sets of drivers required for that board...one for the JMicron SATA-II controller, and another for the ULi Southbridge controller. Sounds like it's the southbridge that's the problem.Make sure you didn't install USB drivers from the disk that came with the mobo...they are for Win 98/ME...and I believe the USB ports are on SouthBridge, so bunged-up drivers there could be a problem. Just use the default WinXP drivers there.Make sure you don't have the SATA ports in RAID mode. And also try changing the spread spectrum settings for the port...and disable the gated clock.Given that you have a good working 300 Gb/s system drive via the JMicron controller, you could always disable the onboard SB SATA controller and put a PCIe or PCI SATA-II interface board into a card slot and use that instead.Good luckBob ScottATP IMEL Gulfstream II-III-IV-VSantiago de Chile

Hi Bob,Thank you VERY much for your comprehensive and helpful reply.To take your points one by one:"You haven't said if this PC is a new build or not...i.e. have the SATA ports worked in the past?"Yes, it's a fairly new build. Originally I was using my old Hitachi Deskstar 80GB (Archives) on the ATA133 MoBo interface and it worked fine with none of the problems now being experienced with my SATA drive. It should be noted that when I installed the WD Caviar SE16 400GB (WD4000KD) SATA I was unable to copy the image across from the Hitachi Drive using Acronis True Image Home 9.0 - same problem, system locked up (this was from within Windows). However, I was able to complete the exercise successfully using the previously created Acronis Bootable Rescue Media CD.When I ran HD Tach's Long Bench (32MB zones) it fails with a system lockup during the sequential read phase.My experience, therefore, with the SATA ports is that they have never worked properly. SATAII, however, is fine."There are actually two sets of drivers required for that board...one for the JMicron SATA-II controller, and another for the ULi Southbridge controller. Sounds like it's the southbridge that's the problem.?"Yes, I'm sure you are correct. In fact, I did not install the chipset drivers from the CD that came with the MoBo. Instead, I followed my usual practice of downloading and installing the latest drivers available from the AsRock Support download page.http://forums.avsim.net/user_files/164320.jpghttp://forums.avsim.net/user_files/164321.jpg"Make sure you don't have the SATA ports in RAID mode."The SATA ports are configured in the Bios as 'non-RAID'. "And also try changing the spread spectrum settings for the port...and disable the gated clock."Now you've got me... could you elaborate a little about changing the spread spectrum settings for the port and disabling the gated clock. These are areas about which I know nothing. What do they do and can you suggest values I should try for the spread spectrum (currently set to 'Auto' and the only other option available is 'Disabled')? Incidentally, I note the 'Gated Clock Function' is already disabled."Make sure you didn't install USB drivers from the disk that came with the mobo...they are for Win 98/ME...and I believe the USB ports are on SouthBridge, so bunged-up drivers there could be a problem. Just use the default WinXP drivers there."Yes, I'm aware that XP fully supports USB 1.1 and 2.0 and no additional drivers are needed unless a 3rd party USB card is installed - not relevant here.http://forums.avsim.net/user_files/164322.jpgRegards,Mike

Spread spectrum, as the term is used w/r/t computer bus design, is basically a feature that floats the bus clock freq around in order to avoid generating an RF spike at a particular frequency...a feature designed to meet RF emissions standards in places where they are more stringent. It can be bad juju if you're overclocking the PC, and generally it's a useless performance drain otherwise. I don't remember you saying if you were trying to OC this machine or not.If none of this works, I'd consider RMAing the board. A quick google search didn't reveal evidence of a rash of similar problems with this board...probably you have a bad southbridge LSI or other faulty component on the mobo. Given it's new, I'd chalk it up to infant mortality and move on.CheersBob ScottATP IMEL Gulfstream II-III-IV-VSantiago de Chile

Hi Bob,Thank you!Your explanation appears to suggest that I should try the effect of disabling 'Spread Spectrum'. Am I correct?No, I am not one for overclocking components, but if I have good quality ram I will try adjusting the timings to eek out a little extra performance.I'll try changing the cable...seems unlikely, but you never know ;)Regards,Mike

Yes, I'd try disabling the SS...and most definitely try a different SATA cable.CheersBob ScottATP IMEL Gulfstream II-III-IV-VSantiago de Chile

Hi Bob,Just thought you would like to know how it all turned out in the end.I am pretty sure I have tried everything to get this to work but, sadly, no joy.For a few hours over the weekend I had convinced myself that one of the background processes and/or services was implicated. I have 2 hardware profiles: One for general use and one for simming and gaming with very little running in the background. While running under the latter profile, HD Tach was completing, as was copy operations of large files, scanning with AVG, etc. So, I started the tedious process of systematic elimination, one-by-one, but the problem (lockups) eventually returned after extended repeated testing. It seemed I was winning when all of a sudden the raspberry was blown and I felt I was back to square one. The behaviour was consistently inconsistent, and therefore could not be relied upon.Time to make a decision as far too much time and effort was being expended :( I decided to ditch SATA and return to the ATA 133 interface. After imaging the backup onto the UDMA-6 320GB Hitachi Deskstar T7K500 7200RPM and resizing the partition to fill the unallocated space, I started testing.This drive hasn't put a foot wrong and HD Tach is returning better results than the SATA drive!Western Digital 4000KDRandom Access: 13.7msCPU Utilization: 7% (+/- 2%)Average read: 55.6 MB/sBurst Speed: 104.1 MB/sHitachi Deskstar T7K500Random Access: 13.1msCPU Utilization: 4% (+/- 2%)Average read: 61.7 MB/sBurst Speed: 123.6 MB/sMy WD2500KS SATAII drive continues to return flawless performance and my system seems quite stable. So, I guess, something on this board is preventing SATA and SATAII from working together. Perhaps the 2xSATA ports would work in RAID, but I ain't gonna be the one to try it!!Thanks again for all your help. I thought about introducing an additional card to provide alternative SATA functionality, but the only place it would go is in the PCI-E slot and I'm reserving that for my next GPU upgrade. The ASRock board is pretty versatile, but the downside is that it can be a bit of a squeeze trying to fit everything in. After you've installed an AGP graphics card, a sound card and additional USB 2.0 ports which block a PCI slot then there isn't much room for anything else.Beat regards,Mike

Hi Mike; Thanks for "the rest of the story." Might be instructive to someone else with that sort of problem. And proof positive there's more than one way to skin a cat.CheersBob ScottATP IMEL Gulfstream II-III-IV-VSantiago de Chile

Hi Bob,I've made a further discovery!When I first installed my 2 memory modules I tested them thoroughly and they passed with flying colours. It never occurred to me to look at them again...until now!I checked them again with Memtest86+ v.1.65I got it started intending to let it run overnight. When I returned in the morning the HD light was on and steady and I could not regain control of my LCD monitor!! The only way out was to power off.Without wishing to labour the point, I eventually found that Memtest86+ was failing on Test 8 (Modulo 20, ones and zeros). As you know, the software allows you to select inidividual tests to loop around and this confirmed there was definitely a problem with this test and I was able to reproduce the behaviour as seen when I returned to the PC yesterday morning. And that ran a few bells ;)I had forgotten that some time back I had been experimenting with the latency values for my Crucial 2x1GB PC3200 DDR400 RAM. The defaults selected by Auto settings in the BIOS are 3-3-3-8 (2T). I had changed the CAS value to 2.5 and the Command Rate to 1T. So, the settings as shown in SiSoftware Sandra Professional Home XI had now become 2.5-3-3-8 (1T). These settings were providing quite a useful boost in memory performance.After further experimentation, I find that the Command Rate value of 1T is the problem. Resetting it to 2T and the Memtest86+ problem at Test 8 goes away. I left Memtest86+ running at the software's default settings for 12+ hours yesterday and no errors were detected.For the time being I have left the CAS Latency value at 2.5I had been experiencing very infrequent random program crashes (not system lockups) in heavyweight software with known previous stability. Included were FS9 and FSX. I am now wondering whether my changing the memory Command Rate was responsible. I suppose it is also possible that this change was causing my SATA problems. I confess that, without knowing how they work, I don't understand why this should be the case unless memory is used extensively during Formatting, CheckDisk and AVG scanning operations and timing is critical. I can understand why the lockups were occurring during disk to disk copy operations of large files as memory would certainly be used and extended read operations are involved. Also, why did HD Tach sometimes complete during the long bench sequential read phase and not fail everytime? Luck? And then there is the SATAII. Why was it unaffected other than it uses different drivers? I would certainly welcome any further comments on these points?At the moment things seem to be okay, so I don't feel inclined to open up the case, yet again, just to retest the SATA drive (you've got to see my desktop layout to understand!). Doubtless I will at some stage in the future.Best regards,Mike

I think the odds are pretty good that this was the root of your SATA problem.Anytime I change RAM settings (voltages or timings) I run MemTest86+ through 5-10 complete cycles, and then I run a long (at least 12 hrs) Orthos test over the next available evening. Have caught a number of weak configurations that were not immediately obvious this way.CheersBob ScottATP IMEL Gulfstream II-III-IV-VSantiago de Chile