| |
ciccior2003
Account closed
Registered: Jun 2012
Posts: 12 |
Ultimate II+ .crt+reu
i have tested a demo made to work in .crt + reu. It works in vice but with 1541 ultimate II+ to work i have to start demo,crt than disable reu from menu. Return to demo, enable the reu and reset the system without reload the demo.crt.
I have made the same with RR.crt and Reu doesn’t work until i don’t disable and enable the Reu.
There is a settings to start .crt and use Reu without this method? |
|
| |
Walt
Registered: May 2004
Posts: 47 |
This might be off-topic, but when developing Expand ( Expand ) I had similar issues. Check the note file on the disk for more details.
It is a REU trackmo and if I did not wait some frames (3-5) between utilizing REU and drive access it would randomly hang.
As you experienced, it worked fine in VICE but would fail on 1541 Ultimate II+.
I had some mailing with Gideon but never received an answer about the underlying issue, so this is purely speculation:
"I suspect the memory on the 1541 Ultimate is used for both the REU and drive emulation, carts etc. and as the REU is pretty demanding cycle-wise it can fail when some functions are used at the same time."
Also the timing on the REU is not 100% correct but in most cases it does not matter. See REUTools for further info/test ( REUTools 1.0 )
Even with stuff like this I still find the 1541 Ultimate II+ an amazing product. And knowing the quirks is first step to work around them :) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Working around emulator bugs should be punished by law though |
| |
hedning
Registered: Mar 2009
Posts: 4595 |
+1. Real hw should be the only benchmark and goal. Then emulators should follow what's working (and not) on real hw. |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
Utilitarists may argue that if some piece of code works on both real hardware and emulators there would be more happiness in the world, all other things being equal, but I'm happy to see that Groepaz and Hednings looks as the issue as A MATTER OF PRINCIPLE!!!! ;D |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
The problem with this is that when software is made to work on broken emulators, noone will feel the need to improve those emulators.
The contrary should be true - break emulators loud and proudly. See what crossbow did with VICE. |
| |
Frantic
Registered: Mar 2003
Posts: 1627 |
I agree |
| |
Raistlin
Registered: Mar 2007
Posts: 552 |
The “product” (the demo) should be fine to work around the poor emulation if it can... as a courtesy, the developer of the product could let the creators of the emulator know about the fault and perhaps provide a test case to them?
If we release something that doesn’t work quite right in VICE, that is easy to fix, it will only lead to downvotes on CSDB due to people not using real hardware ... |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
If you care about votes then you have bigger problems than your product working in an emulator. |
| |
ciccior2003
Account closed
Registered: Jun 2012
Posts: 12 |
I agree with all purists of real hw, but a real Reu is impossible to find and i think that in c64 scene the 95% of people use Vice and less than 5% use Ultimate or other cartridge emulators.
The only way to see decend interlaced effects is to use a real c64. So i think 1541 ii+ and RETROTINK 2X is the best solution price/quality. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:in c64 scene the 95% of people use Vice
in the actual c64 scene 100% use real hardware. |
| |
TheRyk
Registered: Mar 2009
Posts: 2062 |
How about contacting Firmware developers of 1541UII+?
If it's still markus, he'll be glad to help if it's fixable via fw update.
And he's very trustworthy, he fixed some EF issues of 1541UII+ some years ago with a some crack we've never released and never leaked anything :) |
| |
ciccior2003
Account closed
Registered: Jun 2012
Posts: 12 |
Quote: Quote:in c64 scene the 95% of people use Vice
in the actual c64 scene 100% use real hardware.
Can be true but 95% of demos are seen from vice or youtube.. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
the point being? |
| |
ciccior2003
Account closed
Registered: Jun 2012
Posts: 12 |
the point is a compromise to have real c64 and emu cartridge to have simil REU.
I think that using virtual-REU mixed at .crt can boost the c64 in new way. |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
wat |
| |
Raistlin
Registered: Mar 2007
Posts: 552 |
The point is... if we’re creating a product for people to enjoy... and if it doesn’t work in an emulator - and we can easily fix that... we will likely fix it. And let the developer of the emulator know what we found.
I agree that the pure scene is on real hardware. BUT... you can’t on one hand say that the people making releases have a responsibility to release something that bugs on an emulator just to push the developers of the emulator to fix their shit... while also saying that those developers won’t be bothered to fix it if we just tell them about it? That makes it sound like the emu devs don’t care about purity - but that the release (demo/intro/whatever) devs should? |
| |
JackAsser
Registered: Jun 2002
Posts: 1989 |
Quote: The point is... if we’re creating a product for people to enjoy... and if it doesn’t work in an emulator - and we can easily fix that... we will likely fix it. And let the developer of the emulator know what we found.
I agree that the pure scene is on real hardware. BUT... you can’t on one hand say that the people making releases have a responsibility to release something that bugs on an emulator just to push the developers of the emulator to fix their shit... while also saying that those developers won’t be bothered to fix it if we just tell them about it? That makes it sound like the emu devs don’t care about purity - but that the release (demo/intro/whatever) devs should?
I'd never tbh. If it works on the real thing and not in VICE, well, tough luck. I would likely try to fix VICE myself actually instead of spending time working around the bug in the demo/game. Or at the very minimum harass Groepaz and Blacky. |
| |
Walt
Registered: May 2004
Posts: 47 |
I agree, of course, that real hardware is the correct way. And as others have said, finding a real REU is hard.
1541 Ultimate is a great product and as I wrote, I have contacted Gideon but he has not (AFAIK) fixed the issue.
Me being oldschool means that I still use old VICE versions, especially as I'm not very fond of GTK on Windows ;)
If I had released Expand with no checks and no delay for fixing the loader issue on 1541 Ultimate I'm 100% certain that I would be getting the blame :
"He can't even time a magic byte raster effect, what a lamer!!!11"
"The demo crashes at random, when will he learn to code???"
The above was also the reason for releasing REU Tools, not so much in regard to VICE as the newer versions (and dropping x64 in favor of x64sc) means that only the older versions have problems, but then the hardware implementors have something to test their products... |
| |
TheRyk
Registered: Mar 2009
Posts: 2062 |
Walt: I used to think this way, too (if coder can't work around new fancy hardware spoofs, coder is the one to blame). But the firmware developers of both 1541-U-II+ and TC64 assured me, it's the other way round. If sth works on real (ancient) hardware but not on "unreal" but nice-to-have new stuff, it's up to the firmare developers to make their hardware behave as accurately as possible compared to original hardware. |
| |
oziphantom
Registered: Oct 2014
Posts: 478 |
lets do our best to avoid the ZSNES problem ;) |
| |
Krill
Registered: Apr 2002
Posts: 2839 |
Quoting oziphantomlets do our best to avoid the ZSNES problem ;) I believe the C-64 equivalent is the "Amber Cow effect" =) https://csdb.dk/release/?id=4820&show=goof |
| |
Walt
Registered: May 2004
Posts: 47 |
@TheRyk: You are right, but the problem is that if it is just us thinking like that, there is still the risk of blame from those who do not think like us (or do not understand the problem) :(
As I see it, I have found a problem, worked around it and documented it and made test software for those that need to fix it... I hope that this way everybody wins ;)
@Krill: Amber Cow looks like they were able to use graphics under $9xxx in earlier VICE versions? |
| |
Krill
Registered: Apr 2002
Posts: 2839 |
Quoting Walt@Krill: Amber Cow looks like they were able to use graphics under $9xxx in earlier VICE versions? Indeed! Although that's possible using an actual real Action Replay (e.g.), too. =) |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
Quote:That makes it sound like the emu devs don’t care about purity
There are countless examples for this, regarding VICE, regarding 1541U, regarding TC64, and pretty much every other emulator in existance. When things work, there is no reason to fix anything. The only reason for most authors to fix something is when a real program (not some testprogram) breaks. Testprograms are only useful (and needed) for debugging something, incase some real program failed. Infact, not seldomly when i contact other emulator authors about certain failing tests, the first question is "but what does it break?" - and when i cant name some real program they will likely not spend a lot of effort into fixing anything.
As a VICE and TC64 developer, i ask you to break the emulation, don't work around it. You are not doing anyone a favour.
Quote:
"He can't even time a magic byte raster effect, what a lamer!!!11"
"The demo crashes at random, when will he learn to code???"
and the only proper reaction to this is "screw you, it works on real hardware."
PS: for TC64 we actually spent quite a bit of effort into fixing a bunch of recently released REU demos. apparently that they work now doesnt actually mean we fixed it. good work :( |
| |
Walt
Registered: May 2004
Posts: 47 |
@Groepaz: With your test-suite readily available, shouldn't there be competition between emulator developers (both software and hardware kinds) for the 'most accurate' bragging rights, similar to who gets first release on a crack? Or am I missing something??? |
| |
chatGPZ
Registered: Dec 2001
Posts: 11108 |
What should be and what is are two different things. And its always a lot more interesting to add new features to your emulator than fixing some obscure theoretical problem that only exists in some test program. |
| |
ciccior2003
Account closed
Registered: Jun 2012
Posts: 12 |
I posted in github this 1541 uII+ bug, hoping Gideon can resolve it in the next public firmware . |
