|
|
Post by mirkosoft on Apr 17, 2016 18:45:08 GMT
Hi! That's great! 8MHz... Even Z80 in CP/M Cartridge normal clocking 8MHz and again sync with VIC-II causes 4MHz really ~3MHz... your link contain test program! Thank you! Sadly, my opinion (each can have own) is that sadly C128 features C64 mode and this means VIC chip - this is biggest disadvantage of C128... even C64 mode is less compatible for differences between VIC-II and VIC-IIe. Some 3D games (I tested them with SCPU128 on real hardware and xscpu64 of VICE) working on 128 in 64 mode not and on 64 yes (of course I can't compare to real C64 and real SCPU...). I plan this weekend end with capturing video of these games (Brotcasten, Doom 2.5, Mood, Dark Side...) of C128 with SCPU128 and Mega 65. I can post link here. For now disk image with really simple examples and Z80 handler. Miro Attachments:80.d64 (170.75 KB)
|
|
|
|
Post by C128Man on Apr 18, 2016 11:42:31 GMT
Maybe another interesting thing. In the german magazine C't from december 1987, they build an 8 MHz (!) Z80 cartridge for the C64. To benchmark the device they run the program muMATH to calculate the factorial of 293. Results: C64/Z80 (original Commodore cartridge): 64 seconds. C128/Z80: 42 seconds. C64/Z80 (8 MHz cartridge): 34 seconds. I don't really understand. Mirkosoft says that the Z80 if the CP/M cartridge runs at ~3Mhz. We known that the Z80 of the C128 runs at ~2Mhz. But, then benchmark shows that the Z80 of the C128 is faster than the C64. It's not logic  There's something I have not understand |
|
|
|
Post by remark on Apr 18, 2016 13:52:06 GMT
@ C128Man The magazine is not very clear on the test conditions. But my guess is, the integration of Z80 in the C-128 is much better than with the CP/M cartridge. You can see that the 8 MHz cartridge is only 1,24 as fast as the C-128, instead of 2 times as fast (the Z-80 in the C-128 runs at 4 MHz, but can only access memory half of the time). But to be more conclusive we need more benchmarks. |
|
|
|
Post by mirkosoft on Apr 19, 2016 16:04:19 GMT
Hi! Like I said, forgive me that later, I created 2 videos - testing C128DCR with SCPU128 and CP/M cartridge and Mega 65. Comments are inside. To author of Boulder Dash 128: please forgive me that I didn't wrote your name in video, but in video capturing progress I was concentrated to recording and work. I know that at cutting video I could to do it, again, I was so tired that I was not able, video was finalized at 4AM CET... so, please forgive me. Here are links (to download - videos are large file, M65 video was recorded in HD and C128 video similar resolution): C128DCR w/ SCPU and CP/M cartridge testMega 65 testDownloadsSo, to compatibility C128 with C64 - it is really not so bad, but when is used SCPU it's same for both machines (C128 less compatible 'cause C64 is real computer) X65 architecture is replaced by 65K... Compatibility of M65 with C65 and C64 - C65 is compatible more, but many things are not finished (bitplanes etc.) and C64 wil be finally not so incompatible - Paul is working on "CPU switch" for both modes (C65 and C64) between 65CE02 and 6502 incl. illegal opcodes. So, maybe we will see more than is awaiting. Miro 27 |
|
|
|
Post by mirkosoft on Apr 20, 2016 19:03:53 GMT
|
|
|
|
Post by hydrophilic on Apr 26, 2016 6:01:57 GMT
All I can say is wow! I do not own any of the hardware you have, mirkosoft, but thank you for your (many) detailed reports!!! Hopefully this will help other C128 owners who *do* have the hardware!! |
|
|
|
Post by nonefornow on May 1, 2016 15:19:39 GMT
I think it was written by some hardware folks at Commodore (maybe Bil Herd?) that CP/M cartridge would not work with C128 in any mode? Even without SuperCPU! Don't know where I read that or how true it is... Page 12 of the C128D System guide states "If you have the external CP/M 2.2 cartridge marketed for the Commodore 64, do not plug it into the Commodore 128. The Commodore 128 has a Z80 microprocessor already on-board for CP/M 3.0. If you do plug in the CP/M 2.2 cartridge, it can cause unpredictable results."
On the same page later on the guide also states "If you have the 1764 RAM expansion module for the Commodore 64, DO NOT use it with the Commodore 128. The 1700 (128K) or the 1750 (512) RAM expansion modules are designed for the C128 hardware.
I know that is not true. The C1764 cartridge works perfectly well in the C128D.
And it looks as if Miro has proven that also the statement about CP/M is not totally correct.
|
|
|
|
Post by mirkosoft on May 1, 2016 23:50:37 GMT
In all cases - Bil Herd talks about Z80 in C128 like big hack, my test was without high hopes, but I wanted to give back to my C128 Z80 also when is SCPU powered on, later I could expand Z80 support in The Ace.
First, when I connected cartridge into C128, I meant it will redirect C128 to C64 mode, but surprise was no redirection, so I meant that cartridge is in C128 invisible, but I set C64 mode and with small hope inserted CP/M 2.2 disk into drive... I began booting. I meant still that's beginning of boot and it will later crash, like CP/M 3.0 when is SCPU turned on. So, I removed cartridge and tried the same - surprise: CP/M 2.2 boot after entering command RUN only simple returns power-on message - and it was beginning... of many surprises and success... I must note that first tests were with SCPU turned off.
In case REU - also in 1700 and 1750 manual is written that will not work on C64 - it needs higher PSU and works. I'm using REU 1764 expanded to 1,5MB too and without any problem.
Not long time ago I read more about Turbo Chameleon 64 and there is written: "Don't connect to C128! Can damage C128 or CH64 or both!" - but where is true? I'm not rich man and don't need to damage my C128 and Chameleon is FPGA only with closed core...
BTW: Since I began flirt with CP/M, on Commodore 128 used CP/M80 2.2 snd 3.0 (Plus), I have bootdisk of CP/M86 1.1 also for i486 and found disk for ZXS with CP/M80 2.2...
I want to try CP/M68 on my MacSE.
When I compare MacSE with MC-68000 @6mhz with C128 with WDC-65816 @20mhz - Mac is sloooow, I mean if I could compare also A500 can run slower...
Miro
|
|
|
|
Post by nonefornow on May 9, 2016 21:51:44 GMT
but I set C64 mode and with small hope inserted CP/M 2.2 disk into drive... I began booting. I meant still that's beginning of boot and it will later crash, like CP/M 3.0 when is SCPU turned on. So, I removed cartridge and tried the same - surprise: CP/M 2.2 boot after entering command RUN only simple returns power-on message - and it was beginning... of many surprises and success... I must note that first tests were with SCPU turned off. Miro I was wondering if you have a .D81 of that CP/M 2.2 disk. |
|
|
|
Post by mirkosoft on May 9, 2016 22:51:18 GMT
No, I have only D64 and real 5,25" DSDD floppy and many others CP/M images (CP/M65, CP/M86...etc.), but later I can do any other, it's simple, currently I'm tasks overdosed...
Miro
|
|
