VOGONS


First post, by Paar

User metadata
Rank Member
Rank
Member

I tried to tinker with my 486 motherboard again (Chicony CH-471B) and managed to finally get everything to work properly. Apart from one thing - I have speed issues regarding integrated VLB IDE controller (in my case Appian AD/2). I use IDE to SD adapter in my system with fast SD card (20MB/s write speed in theory) and when used with this integrated controller I'm getting around 1500kB/s reading speed. That's rather low as I can get the same speed with the integrated ISA IDE controller. Tried to change some settings in BIOS and after chaning the system memory from "non-cacheable" to "cacheable" the speed raised to around 1600kB/s. Better but still not great.

Tried my Winbond VLB IDE card and with that I got around 2500kB/s reading speed which is much better (the old hardware still limits the potential peformance of the SD adapter but I'm fine with that). The question is why is the Appian controller so much slower? Am I doing something wrong or is it just piece of junk?

I'm including picture of the Speedsys results for the Appian controller. Thanks for any tips!

V98COsAt.jpg GCXYdxvt.jpg

Reply 1 of 6, by mpe

User metadata
Rank Oldbie
Rank
Oldbie

You usually get better performance with CF cards as they work natively on the PATA bus. IDE to SD brings in another protocol translation layer which can limit performance. Not sure if that is the problem though.

In order to get higher speeds you need to configure the IDE controller to use higher PIO speeds. This is usually done by jumpers on the motherboard or by a driver. If you have no such configuration most likely you are using a conservative setting like PIO0 which limits the performance.

Blog|NexGen 586|S4

Reply 2 of 6, by kixs

User metadata
Rank l33t
Rank
l33t

Paar:
I've noticed the same thing when tested different VLB I/O cards. But I didn't have the motivation to do further investigation (jumpers, other motherboard, CF card or adapter...). Even ISA I/O cards can manage up to 2000KB/s transfer speed.

Requests are also possible... /msg kixs

Reply 3 of 6, by Paar

User metadata
Rank Member
Rank
Member

As there are no jumpers on the motherboard for this (according this link) I will probably have to use some driver, if I find any. Or I could try the DOS version of PowerIDE utility. Hopefully I'll make it work.

Reply 4 of 6, by mpe

User metadata
Rank Oldbie
Rank
Oldbie

Yes. This is an early IDE controller. So drivers won't be talking about PIO/DMA modes. Most likely you'll be able to change active/recovery clock cycles.

Here is the spec. So I believe some tools for programming these registers must exist:

http://www.ryston.cz/petr/vlb/adi2.html

Depending on your BIOS version you might also try to update your BIOS or try to reveal hidden settings that would allow you to set IDE controller registers

Blog|NexGen 586|S4

Reply 5 of 6, by Paar

User metadata
Rank Member
Rank
Member

I have contacted a member of VCFED forum who claims to have proper drivers. And I will try the PowerIDE app, it's worth a try. I will share my results here. I hope I will be able to bring the board to it's full potential.

Reply 6 of 6, by Paar

User metadata
Rank Member
Rank
Member

Success! After some digging on the internet I've managed to secure DOS/Win3.1 drivers for the controller. Installed the DOS one (which is one *.SYS file) and after that the reading speeds increased dramaticaly. I think the only thing that limits the SD card is an old IDE standard with max 8 MB/s transfer rate. Thank you mpe for a tip, it really helped.

StY4Stgt.jpg

I'm including the driver for anybody who would find it useful.

Attachments

  • Filename
    adi2drv.zip
    File size
    108.04 KiB
    Downloads
    38 downloads
    File comment
    Appian AD/2 IDE Controller DOS/Win3.1 Drivers
    File license
    Public domain