TT Rom access speedup

The latest tweaks and fixes to keep your machine running in tip-top shape.
Atarian Computing
Posts: 444
Joined: Tue Aug 22, 2017 4:27 am

Re: TT Rom access speedup

Post by Atarian Computing »

guus.assmann wrote: Thu Mar 07, 2019 5:41 pm Hello Josh,

The Eproms from ThunderStorm are fast enough, mine are 150nS.
And 1N5819 should work, most important is the voltage drop. ( <0,8V)
However, CS is pin 20 and not pin 22.
Pin 22 is OE and that's permanently connected to GND trough one of the jumpers.
I have not seen any slow Eproms yet. But I've only tested on 3 TT's.
And I've used Germanium diodes. I could test with 1N5819 as well.

BR/
Guus

PS: Attached are the GB files with the speed data.
Ok, I'm confused. 27c010 pin 24 is OE and 27c1000 pin 2 is OE. Both have pin 22 as CS. Now 27c512 has pin 20 as CS and 22 as OE. 28pin 1mbit Mask rom has pin 20 as CS. How are your W601,W602,W603 jumpers set? Maybe you're using different eprom chips?
tt-roms.jpg
tt-roms.jpg (175.36 KiB) Viewed 6020 times
Steve
Posts: 2570
Joined: Fri Sep 15, 2017 11:49 am

Re: TT Rom access speedup

Post by Steve »

Once Atarian has an easy to follow guide I will do this in my TT as well :) And maybe the FPU overclock too.
guus.assmann
Posts: 94
Joined: Sat Sep 23, 2017 8:52 pm
Location: Ede Netherlands

TT Rom access speedup

Post by guus.assmann »

Hello,

I am very sorry for the mixup I caused.
I confused 28 pin and 32 pin devices.
So it is indeed pin 24 that's the CE for the Eprom that I used.
I'll try with 1N5819 diodes, hopefully today.

As for the Gembench, I took the most recent version I could find.
And next I'll try without any extra programs loaded. (Rename the Auto-folder should do the trick)

BR/
Guus
Atarian Computing
Posts: 444
Joined: Tue Aug 22, 2017 4:27 am

Re: TT Rom access speedup

Post by Atarian Computing »

guus.assmann wrote: Fri Mar 08, 2019 8:01 am Hello,

I am very sorry for the mixup I caused.
I confused 28 pin and 32 pin devices.
So it is indeed pin 24 that's the CE for the Eprom that I used.
I'll try with 1N5819 diodes, hopefully today.

As for the Gembench, I took the most recent version I could find.
And next I'll try without any extra programs loaded. (Rename the Auto-folder should do the trick)

BR/
Guus
No worries. Thanks for clearing it up. Although a slight confusion remains after your correction. Have you meant OE (Output Enable) all this time rather than CS/CE (Chip Select/Enable)? Because pin 24 is OE.

Josh.
guus.assmann
Posts: 94
Joined: Sat Sep 23, 2017 8:52 pm
Location: Ede Netherlands

TT Rom access speedup

Post by guus.assmann »

In a TT with Eproms, the OE line will be jumpered to GND, so OE is not used in the selection of the Eprom.
And it does need CS to work.
I have been looking at the data sheet of the 1N58189 but cannot find a recovery time.
That may be an important factor.
I have already had the TT not booting when I had one Germanium diode and a BAT48 (Si).
Unfortunatelly, the Germanium diodes I used had no partnumber.

I will do some more tests this weekend.

BR/
Guus
Atarian Computing
Posts: 444
Joined: Tue Aug 22, 2017 4:27 am

Re: TT Rom access speedup

Post by Atarian Computing »

guus.assmann wrote: Fri Mar 08, 2019 10:12 am In a TT with Eproms, the OE line will be jumpered to GND, so OE is not used in the selection of the Eprom.
And it does need CS to work.
I have been looking at the data sheet of the 1N58189 but cannot find a recovery time.
That may be an important factor.
I have already had the TT not booting when I had one Germanium diode and a BAT48 (Si).
Unfortunatelly, the Germanium diodes I used had no partnumber.

I will do some more tests this weekend.

BR/
Guus
Ok, thanks again. I measured the voltage drop on both 1N5819s to be 0.22v. So should be fine if germaniums work. So yeah, it must be something other than switching speed.

Josh.
Atarian Computing
Posts: 444
Joined: Tue Aug 22, 2017 4:27 am

Re: TT Rom access speedup

Post by Atarian Computing »

Ok guys. It was the ROM speed. I created a new set of ThunderStorm roms on 100ns chips. Took awhile because I changed the keyboard table.

Happy to announce that the speedup mod works great now! Here are some shots.

This is before the mod on new chips:
Before.gif
Before.gif (12.76 KiB) Viewed 6156 times

This is after the mod on new chips:
After.gif
After.gif (12.45 KiB) Viewed 6156 times

This is "after vs before":
After vs Before.gif
After vs Before.gif (12.19 KiB) Viewed 6156 times

And this is "after vs before (old chips) with NVDI":
After vs Before NVDI.gif
After vs Before NVDI.gif (12.73 KiB) Viewed 6156 times

Quite surprising results and gains it seems. Nice and simple mod. Thanks guus!! :dualthumbup:
guus.assmann
Posts: 94
Joined: Sat Sep 23, 2017 8:52 pm
Location: Ede Netherlands

TT Rom access speedup

Post by guus.assmann »

Just now I've also tested with the 1N5819 diodes. And that also works just fine.
Then I removed the CF-cards so the TT would boot from floppy.
Attached are the result files.
Quite similar to the pictures in previous post.
I've only looked at the Rom speed while I compared the before and after.

The access to the Rom has influence on many results.
It's some measure of how often the Roms are accessed for the various functions.

U116 has an influence on the final speed as it's clocked with 16Mhz to pass on the DTACK signals.
My Epoms are all 120nS (2 TT's) and I've also got 90nS Flash Roms in a 3rd one.

P.S: How did you make the nice Gif's?

P.P.S: Do I see correctly that the results with NVDI are a little slower?
This would mean that the TT-Ram in the system is slower than the Rom.
Attachments
SLOW.TXT
(1.29 KiB) Downloaded 231 times
FAST.TXT
(1.29 KiB) Downloaded 227 times
Atarian Computing
Posts: 444
Joined: Tue Aug 22, 2017 4:27 am

Re: TT Rom access speedup

Post by Atarian Computing »

guus.assmann wrote: Fri Mar 08, 2019 5:04 pm Just now I've also tested with the 1N5819 diodes. And that also works just fine.
Then I removed the CF-cards so the TT would boot from floppy.
Attached are the result files.
Quite similar to the pictures in previous post.
I've only looked at the Rom speed while I compared the before and after.

The access to the Rom has influence on many results.
It's some measure of how often the Roms are accessed for the various functions.

U116 has an influence on the final speed as it's clocked with 16Mhz to pass on the DTACK signals.
My Epoms are all 120nS (2 TT's) and I've also got 90nS Flash Roms in a 3rd one.

P.S: How did you make the nice Gif's?

P.P.S: Do I see correctly that the results with NVDI are a little slower?
This would mean that the TT-Ram in the system is slower than the Rom.
JML-Snap is what I use. Released as freeware by the author. See attached.

It's an NVDI result compared to an NVDI result. What you're seeing (I assume) is a smaller gain than without NVDI.
Attachments
snap_522.rar
(46.21 KiB) Downloaded 242 times
User avatar
Icky
Site Admin
Site Admin
Posts: 3986
Joined: Sun Sep 03, 2017 10:57 am
Location: UK

Re: TT Rom access speedup

Post by Icky »

I need to do this on my TT too. Finding time is the problem.

Thanks for sharing guss :thumbup:
Post Reply

Return to “TWEAKS & FIXES”