exxos's Atari, Amiga & retro forum

PhilC wrote: 29 Feb 2020 08:19 This sounds interesting. Was there an upgrade fee from v10 to v11?

€10 @PhilC

Got mine downloaded this morning.

Please let me know if it's worth the upgrade

Yes, it's €10

I tested HDDRIVER 11 with XFRRATE yesterday afternoon om my Mega 4.

Set up was as follows:

Mega ST 4
16GB CF card pre configured with HDDRIVER 10.18, then upgraded to V11.
Beta ST-IDE-ROM board configured to use TOS 2.06.

I ran tests with Blitter support off, and then enabled it. The results were odd.

Blitter off:


Blitter on:


As you can see with the Blitter on it looks weird, much slower than with it off.

It has occurred to me that I should double check to ensure the Blitter was enabled in TOS, which I'll do later (I'm pretty sure it was), but those results do look a bit strange.

From a user perspective it was difficult to feel a difference.

I'll update a bit later.

Thanks @rubber_jonnie , I guess I'll wait for now before upgrading, especially considering I only got 10 last year.

For comparison. This is on my machine, with Lightning ST.

HDDRIVER 10: HDDRIVER 11 + Blitter enabled: Be aware that this is with a crappy SD-IDE-adapter that does not support "R/W multiple" ("Accelerated Mode" in HDDRIVER); otherwise the speed increase from version 10 to 11 would be bigger, as I've seen during my beta tests.

PhilC wrote: 01 Mar 2020 08:51 Thanks @rubber_jonnie , I guess I'll wait for now before upgrading, especially considering I only got 10 last year.

@PhilC I just went back and cross checked my results to make sure Blitter was enabled in TOS and I get exactly the same results. I'll probably have another look later today, and may drop Uwe a note, because it could be that XFRRATE is getting a weird read because of the Blitter. It certainly doesn't seem to be that much slower, at the very least the same speed.

@exxos do you know of any transfer rate testing tools aside from XFRRATE?

rubber_jonnie wrote: 01 Mar 2020 09:35 @exxos do you know of any transfer rate testing tools aside from XFRRATE?

I don't know of any others. Though I half think its a limitation of the IDE design. It was never intended to be used in any way other than a basic IDE test board.its by no means a final polished product. I think it will behave fine on the V1.1 which icky is going to build up and test soon.

exxos wrote: 01 Mar 2020 11:41

rubber_jonnie wrote: 01 Mar 2020 09:35 @exxos do you know of any transfer rate testing tools aside from XFRRATE?

I don't know of any others. Though I half think its a limitation of the IDE design. It was never intended to be used in any way other than a basic IDE test board.its by no means a final polished product. I think it will behave fine on the V1.1 which icky is going to build up and test soon.

@exxos Yeah, I think perhaps the results from testing one single IDE interface aren't sufficiently diverse to make any kind of call. Hopefully I can get my hands on a V1.1 board and retry, see what the results are.

Honestly, I don't believe the XFFRATE results, because performance is certainly not noticeably worse with the Blitter options enabled.

I should also try it with a genuine hard drive to see if it make any difference over the CF device.