When testing my border art viewer, I recently had a couple of bug reports from people using 48K ZX Spectrums with ULA7.
Has anyone ever run tests on ULA memory access patterns for different versions of ULA? Does ULA7 behaves identically to earlier versions, or is it somehow different?
It would also be great to know if you've ever noticed differences in multicolour effects across different revisions of ULA. I always assumed that revisions of 48K ULA stayed fundamentally identical for timing purposes, but maybe I was wrong.
Timing differences between different versions of 48K ULAs?
-
1024MAK
- Bugaboo
- Posts: 3123
- Joined: Wed Nov 15, 2017 2:52 pm
- Location: Sunny Somerset in the U.K. in Europe
Re: Timing differences between different versions of 48K ULAs?
They did improve the timing of the RAM accesses for later boards and ULAs, but that should not affect a program normally.
Of course, it may be that when combined with the slight differences in timing when a machine has 'warmed up', it may affect border effects or other timing sensitive ULA video tricks.
Some issue one and some early issue two boards have extra daughter boards to correct I/O contention issues. This was then incorporated in later issue ULAs. Some may have escaped modification...
And lastly, later ULAs used a bigger, improved chip compared to earlier ULAs.
Oh, and although there are not many, (if any) boards out there, Sinclair did apparently try a different ULA manufacturer.
Mark
Of course, it may be that when combined with the slight differences in timing when a machine has 'warmed up', it may affect border effects or other timing sensitive ULA video tricks.
Some issue one and some early issue two boards have extra daughter boards to correct I/O contention issues. This was then incorporated in later issue ULAs. Some may have escaped modification...
And lastly, later ULAs used a bigger, improved chip compared to earlier ULAs.
Oh, and although there are not many, (if any) boards out there, Sinclair did apparently try a different ULA manufacturer.
Mark
Standby alert “There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Looking forward to summer later in the year.
Re: Timing differences between different versions of 48K ULAs?
I know that there were multiple revisions of ULA and that some Spectrums timings shifted slightly wrt to frame interrupt as they get warmer. The latter effect is incorporated into most emulators and seems well understood. However, the former has never been explicitly verified.
The easiest way to verify would be to run the old mtest, ulatest3 etc routines by Jan Bobrowski (see e.g. https://softspectrum48.weebly.com/notes ... -challenge), while controlling for the ULA version. If variations of ULA can have slight differences in timings, some of these tests are very likely to pick this up.
The easiest way to verify would be to run the old mtest, ulatest3 etc routines by Jan Bobrowski (see e.g. https://softspectrum48.weebly.com/notes ... -challenge), while controlling for the ULA version. If variations of ULA can have slight differences in timings, some of these tests are very likely to pick this up.
Re: Timing differences between different versions of 48K ULAs?
Is that what's going on here? I noticed that auction yesterday, and was rather surprised to see an uncased issue 2 motherboard go for over £100. That little board on the CPU is one of these early modifications?
Derek Fountain, author of the ZX Spectrum C Programmer's Getting Started Guide and various open source games, hardware and other projects, including an IF1 and ZX Microdrive emulator.
-
1024MAK
- Bugaboo
- Posts: 3123
- Joined: Wed Nov 15, 2017 2:52 pm
- Location: Sunny Somerset in the U.K. in Europe
Re: Timing differences between different versions of 48K ULAs?
Yes. It's known as the dead bug modification.
Mark
Standby alert “There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Looking forward to summer later in the year.
Re: Timing differences between different versions of 48K ULAs?
Yes, I got that one, since I was missing the so-called 'Issue 1.5' motherboard in my collection 
They are quite rare, possibly even more than Issue 1 ones. Now if I only could put my greedy hands on an Issue 5 (something which I know will never happen...)
They are quite rare, possibly even more than Issue 1 ones. Now if I only could put my greedy hands on an Issue 5 (something which I know will never happen...)