How to connect SSD to SATA1 (SandForce based SSD). And a small dirty test.

Alright so in order to try that I need to first install Windows through a standard IDE HDD, just to use the program. I was hoping it was doable in bios in someway..

Everyone was hoping VIA would release a new firmware version for the 8237 chipset so everyone could get their brand new SATA 2 hard drives working.
But instead, VIA released a new version of the chip.

If VIA does not see the drive in BIOS, the operating system will not help.
If you're not looking for adventure, buy a SATA-IDE bridge adapter, for example, Jmicron based.

Hi,

I also have A8V Deluxe...
I have better luck with Red Sata ports, but you will need to load the drivers before windows install.

Using the red sata ports, during bios /startup you press ctrl+f to get into raid menu.

You can then, define a Array for it (button 1).
The computer will do all the Array stuff for you, just need to press Yes's

Do note: You will lose data on the SSD by doing this.

---

For your USB issue... Please check the bios and make sure under boot -> hard drives, that your USB is selected.

---
For updating bios : Asus has a built-in software in bios that allows you to flash bios using floppy disk.
(i think combo is ALT + F4, but RTFM because I forget...) You can try that if USB does not work.

You didn't mention the OS you want to use - with Win98, connect the SSD to the Promise RAID controller, install Windows then install the SATA driver along with the rest. With WinXP you can do the F6 routine, but if you feel like getting fancy you can use Nlite and include the driver in the WinXP installation disk.

000557A wrote on 2025-11-08, 19:06:

Hi, […]
Show full quote

Hi,

I also have A8V Deluxe...
I have better luck with Red Sata ports, but you will need to load the drivers before windows install.

Using the red sata ports, during bios /startup you press ctrl+f to get into raid menu.

You can then, define a Array for it (button 1).
The computer will do all the Array stuff for you, just need to press Yes's

Do note: You will lose data on the SSD by doing this.

The advantage of such controllers is that they are already on the motherboard.
And that's all

Disadvantages
This is a full-fledged PCI controller.
- It used interrupts and other resources, and may conflict with other devices.
- A driver is required during installation. Either integrate the driver into the Windows installation image or find a working (!) FDD and a working (!) floppy disk.
- Incredibly poor performance, limited by the PCI bus bandwidth (which is shared among all devices connected to it).
In modern terminology, this is a controller for "cold data", such as an archive of TV series.

And the IDE-SATA bridge - well, you need to buy it, that's true.

I also have an A8V Deluxe. I threw in a separate PCI SATA controller because the southbridge ports do not seem to support SATA 3.0Gbps drives, and the built-in Promise controller requires drivers and creation of an "array" even for a single drive.

I did modify the latest (beta) BIOS to install the latest option ROMs for both the VIA southbridge SATA ports and the Promise RAID controller, but this was after having switched to the separate PCI SATA controller. I have no idea if the latest VIA SATA option ROM resolves the aforementioned compatibility issue with newer SATA drives. Does anyone want me to upload it for testing?

Towncivilian wrote on 2025-11-09, 13:27:

and the Promise RAID controller

If it is not too much trouble, please publish the AS SDD benchmark results on this controller.

Not an apples to apples comparison, but here are the ATTO results on my Asus K8V SE (Socket 754) with a Silicon Power SSD, Win98SE using both the Promise controller and the PATA port with a SATA to PATA adapter. I was expecting more of a performance difference - the PATA port has the edge in write speed, but it is really too close to call. I stayed with the SATA to PATA adapter since the Promise RAID screen slows down the boot up.

The Promise controller is significantly slower than IDE, even on your screens.
It seems to be slightly faster than Silicon Image, but it's on the same level - ‘for cold data’.

In my opinion, Silicon Image/Promise provide ~UDMA80 performance, i.e. faster than UDMA66.
But they fall short of UDMA100.
Naturally, when using one drive per controller, simultaneous operation of two drives (e.g., copying) will be catastrophically worse.
VIA 8237 has UDMA133, so using these controllers is not particularly sensible.
They are good enough on SuperSocket 7/early Socket 370 systems (with UDMA33/66).

shevalier wrote on 2025-11-09, 13:55:

Towncivilian wrote on 2025-11-09, 13:27:

and the Promise RAID controller

If it is not too much trouble, please publish the AS SDD benchmark results on this controller.

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were available. I am trying to connect a SATA SSD to this controller benchmark, but I think it's fighting with my other RAID controller (RocketRAID 1740) and Windows cannot initialize the RAID controller. When I try to use the Promise controller in IDE emulation mode, the system will not boot despite the boot order being set correctly.

However, I did indeed update the VT6240 RAID Controller option ROM from version V2.01 to V4.97.237. I can benchmark this if you wish.

I also updated the Marvell PXE OROM from v1.15 beta 01 (20040607) to v6.60.2.3 (20080422) but that's not really relevant to this thread.

Towncivilian wrote on 2025-11-09, 16:24:

However, I did indeed update the VT6240 RAID Controller option ROM from version V2.01 to V4.97.237. I can benchmark this if you wish.

The latest version I found is "VIA VT6420 RAID Controller VIA VT6420 Serial ATA RAID BIOS Setting Utility V4.93".
Please post the PCI OROM itself, if you don't mind.
I was just very surprised that the IDE-SATA bridge based on Jmicron shows much better results than Silicon Image on the PCI bus.
I wonder how bad Promise is.
Marvel is also interesting, but I've never come across them.

PS I found where 4.97 can be extracted.
https://www.gigabyte.com/ua/Motherboard/GA-VM … v-10/support#dl

Towncivilian wrote on 2025-11-09, 16:24:

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were avail […]
Show full quote

shevalier wrote on 2025-11-09, 13:55:

Towncivilian wrote on 2025-11-09, 13:27:

and the Promise RAID controller

If it is not too much trouble, please publish the AS SDD benchmark results on this controller.

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were available. I am trying to connect a SATA SSD to this controller benchmark, but I think it's fighting with my other RAID controller (RocketRAID 1740) and Windows cannot initialize the RAID controller. When I try to use the Promise controller in IDE emulation mode, the system will not boot despite the boot order being set correctly.

However, I did indeed update the VT6240 RAID Controller option ROM from version V2.01 to V4.97.237. I can benchmark this if you wish.

I also updated the Marvell PXE OROM from v1.15 beta 01 (20040607) to v6.60.2.3 (20080422) but that's not really relevant to this thread.

Hi, could I ask what bios version you performed the updates on? I am interested in flashing it, provided that it's not bios 1017 /1018.

000557A wrote on 2025-11-09, 20:56:

Towncivilian wrote on 2025-11-09, 16:24:

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were avail […]
Show full quote

shevalier wrote on 2025-11-09, 13:55:

If it is not too much trouble, please publish the AS SDD benchmark results on this controller.

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were available. I am trying to connect a SATA SSD to this controller benchmark, but I think it's fighting with my other RAID controller (RocketRAID 1740) and Windows cannot initialize the RAID controller. When I try to use the Promise controller in IDE emulation mode, the system will not boot despite the boot order being set correctly.

However, I did indeed update the VT6240 RAID Controller option ROM from version V2.01 to V4.97.237. I can benchmark this if you wish.

I also updated the Marvell PXE OROM from v1.15 beta 01 (20040607) to v6.60.2.3 (20080422) but that's not really relevant to this thread.

Hi, could I ask what bios version you performed the updates on? I am interested in flashing it, provided that it's not bios 1017 /1018.

Version 1018.002 beta.

Is another version preferential on this board?

I've attached the VT6240 v4.97 OROM here too.

Towncivilian wrote on 2025-11-09, 23:06:

Version 1018.002 beta. […]
Show full quote

000557A wrote on 2025-11-09, 20:56:

Towncivilian wrote on 2025-11-09, 16:24:

I misremembered earlier. I made no changes to the RAID/SATA Promise option ROMs in the BIOS because no newer versions were available. I am trying to connect a SATA SSD to this controller benchmark, but I think it's fighting with my other RAID controller (RocketRAID 1740) and Windows cannot initialize the RAID controller. When I try to use the Promise controller in IDE emulation mode, the system will not boot despite the boot order being set correctly.

However, I did indeed update the VT6240 RAID Controller option ROM from version V2.01 to V4.97.237. I can benchmark this if you wish.

I also updated the Marvell PXE OROM from v1.15 beta 01 (20040607) to v6.60.2.3 (20080422) but that's not really relevant to this thread.

Hi, could I ask what bios version you performed the updates on? I am interested in flashing it, provided that it's not bios 1017 /1018.

Version 1018.002 beta.

Is another version preferential on this board?

I've attached the VT6240 v4.97 OROM here too.

I personally prefer 1015.002 Beta, because it doesn't have the "USB Overcurrent Detected" bugs caused by RAM (8G of ECC will cause the error).

Thanks for the OROM though! Perhaps I can implement it into 1015 when I am free.

Thanks!

Towncivilian wrote on 2025-11-09, 23:06:

I've attached the VT6240 v4.97 OROM here too.

As expected for a minor version, the changes are purely cosmetic in terms of performance.
Unfortunately, we will never know the changelog.

000557A wrote on 2025-11-10, 02:32:

I personally prefer 1015.002 Beta, because it doesn't have the "USB Overcurrent Detected" bugs caused by RAM (8G of ECC will cau […]
Show full quote

Towncivilian wrote on 2025-11-09, 23:06:

Version 1018.002 beta. […]
Show full quote

000557A wrote on 2025-11-09, 20:56:

Hi, could I ask what bios version you performed the updates on? I am interested in flashing it, provided that it's not bios 1017 /1018.

Version 1018.002 beta.

Is another version preferential on this board?

I've attached the VT6240 v4.97 OROM here too.

I personally prefer 1015.002 Beta, because it doesn't have the "USB Overcurrent Detected" bugs caused by RAM (8G of ECC will cause the error).

Thanks for the OROM though! Perhaps I can implement it into 1015 when I am free.

Thanks!

Very interesting. I have encountered that glitch, even with "just" 4GB of RAM, a few times and it is aggravating. I will give 1015.002 beta a try.

Thanks for testing the OROM, shevalier!

Towncivilian wrote on 2025-11-10, 13:35:

Thanks for testing the OROM, shevalier!

There is only one problem with VIA SATA OROM: it does not exist.
Configuration of controller takes place in the main BIOS itself.
This is OROM for the controller when it is in RAID mode (even without creating the RAID array itself).

I flashed a (modified) BIOS 1015.002 and I'm able to consistently use the 4GB of RAM I purchased for this build. Thank you, 000557A!

I modified it with the same latest versions of the Marvell Yukon PXE option ROM and the VIA VT6240 SATA RAID OROM. I also deleted unnecessary microcode - the BIOS includes microcode for Socket 754 CPUs and also some Socket 940s.

I am still unable to test the onboard Promise SATA RAID controller regardless of the combination of settings I attempt. The controller is detected by Windows but cannot start after installing the driver, in either RAID or IDE mode. I think I will need to remove the Highpoint RocketRAID 1740 card to see if that makes a difference, but then I won't have an OS to boot from. I'll fool with it more another time.

Would you like a benchmark of the RocketRAID 1740 controller, shevalier? I believe it uses a Marvell chip.