MiSTer FPGA Forum

zomgugoff wrote: ↑Thu Jun 04, 2020 7:22 am

AmigoMexicano wrote: ↑Thu Jun 04, 2020 1:32 am I've been told that the boot.rom file goes inside: /media/fat/PC8801mk2SR

I still can't use this core, until my SDRAM board arrives, so I can't check this for sure.

So, where does it go?

That is the correct location

OK, my SDRAM board finally arrived, and I can confirm, that is the correct location, everything works like a charm!

@Artemio
I didn't know this.
The circuit equivalent to my uPD765 is made by hard wired so that the same operation is performed based on the specifications, and Micro code is not used.

AmigoMexicano wrote: ↑Fri Jun 05, 2020 10:47 pm

zomgugoff wrote: ↑Thu Jun 04, 2020 7:22 am

AmigoMexicano wrote: ↑Thu Jun 04, 2020 1:32 am I've been told that the boot.rom file goes inside: /media/fat/PC8801mk2SR

I still can't use this core, until my SDRAM board arrives, so I can't check this for sure.

So, where does it go?

That is the correct location

OK, my SDRAM board finally arrived, and I can confirm, that is the correct location, everything works like a charm!

The correct location is also '/media/fat/bootrom/PC8801mk2SR.rom' (just rename 'boot.rom')

hi

perhaps can you take an eye on amstrad cpc 765 component to merge to a good 765 one

I don't know the architecture of amstrad cpc. Is the uPD765 core of amstrad cpc free to use various sector sizes? The PC-88 uses the maximum disk capacity, or for reasons such as preventing illegal copying and other reasons, the sector size changes from track to track, the number of sectors changes, and the single density/double density changes. The D88 format is a flexible format that can support this, and my core Disk emulator also supports it.

@puu

yes amstrad can use sector size 0 to 6 (128 to 8192 bytes) with only 6304 max used per track

generaly it's 9 sectors x 512 bytes (size 2) = 4.5 KB per track x 40 tracks = 180 KB (178 KB free with 2 KB reserved for FAT)

Are all disk write operations delayed? I tried to create a user disk for Zeliard, but it kept failing part way through. I checked the disk emu screen (F11) and saw the red progress bar start shortly after that.

All FDDs are written to SDRAM once and then delayed back to the image. If the red progress bar goes to the far right, the write back is complete, but if there is another write during the write back, the write back should be repeated again.
From the FDC's point of view, writing to SDRAM is always performed normally whether it is writing back or not.
If you are worried, select EJECT to write back again.

puu wrote: ↑Tue Jun 09, 2020 3:38 pm All FDDs are written to SDRAM once and then delayed back to the image. If the red progress bar goes to the far right, the write back is complete, but if there is another write during the write back, the write back should be repeated again.
From the FDC's point of view, writing to SDRAM is always performed normally whether it is writing back or not.
If you are worried, select EJECT to write back again.

I wonder why the write operation is failing... Can anyone list other games that have user disk creation?

Did you use an existing image for the user disk? Did you create a new one?
What kind of media type (2D/2DD/2HD) did you choose when newly created? In the era of PC-88, the mainstream is 2D, so I think there is no problem if you choose 2D unless otherwise specified. FDC is doing PIO transfer with sub CPU, but I think that it may not work because Z80 is so hard for 2HD 's 500kbps transmit speed.

puu wrote: ↑Wed Jun 10, 2020 3:57 am Did you use an existing image for the user disk? Did you create a new one?
What kind of media type (2D/2DD/2HD) did you choose when newly created? In the era of PC-88, the mainstream is 2D, so I think there is no problem if you choose 2D unless otherwise specified. FDC is doing PIO transfer with sub CPU, but I think that it may not work because Z80 is so hard for 2HD 's 500kbps transmit speed.

I tried new images of each type. They all failed.

@Puu

hello

I saw on your site that there was a new version of the core, but the download link is not functional

The link is working now.
Thank you Puu for this new release of PC88 without the need of the 2nd SD Card!

Sorry, since the size of the uploaded file was unusually large, I confirmed that a huge .stp file for debugging was attached, so I deleted it and re-uploaded it.

thank you Puu !!!

Hello Mr. Puu,
I saw the new version of the core pc8801 on your site. It is impossible to compile it

hello

I will use the missing file from the previous version! It will be good ?

it seems to be good !!!

Sorry I forget. Re-uploded in my blog.

Thank you for the reupload !

Hi !
Puu ! I have synthesized the last core and everything seems to work with internal SD, but the characters do not appear on the screen, just the cursor or the underlined_cursor, no chars, when in the previous version everything was seen and worked correctly.

My pc8801mk2sr.rom is 393,2 kB (393216 bytes)

What should be the correct order of the ROMs, for it to work in this new version?

Thanks a lot in advance !

Starting a few versions ago, the font need to be bundled inside the rom. There was a discussion about it that you can check on page 2.

@LeftEmpty. I have been reading the thread patiently but finally I am not clear on the question of the font file. I have been doing tests with very discreet results.
It would be interesting to clarify as far as possible how to cook the final ROM and the font.rom issue, I still do not understand how with the previous version it worked correctly.

Anyway, Thanks very much for your help.

The font.rom found around is 6KB and is, I suspect, a stack of the two files Puu uses for his rom file.
Splitting the file manually (using an hex editor for instance, at address #0800) should do the trick. To be sure, once you're done with your electronic butchery, you can check whether the font.rom is 2048 bytes long and the graphfont.bin is 4096 bytes long.
I haven't built my own new pc88.rom to test yet, so this is just theory!

Hi ! there's no way.
It does not matter what you do, the result is the same as I put in the images of the post above.

Does anyone work the new version correctly?

The .d88 floppies works fine from internal SD, and graphics demos are working fine, video and audio OK. But no chars at the system prompts.

6kB font.rom contains 2kB 8x8dot font and 4kB 8x16dot font in this order.
Therefore, it is necessary to remove the first 2kB.
The quickest way is to delete addresses 0 to 7ff with a binary editor, but if linux can be used
split -b 2048 font.rom
Since it is divided into xaa, xab, xac every 2kB,
cat xab xac >font.bin
I think you can do it by connecting the two back with.

Hi puu
You are very kind responding. I have not been able to find the famous 6KB font.rom and therefore I am still where I was. I give up.

In the end no one has answered my questions and if someone have the rom currently working. It was hard for me to find and put together the ROM for the previous version of the core, but if it is going to be so difficult I will leave the core for a while until the evil ROM appears.

Let's take a look. This is my last attempt.
-It is a real shame that such a small file ruins the entire core experience.If it cannot be used it is as if it does not exist, this core serves as a nice paperweight since nothing can be done., I just wanted to continue testing the core, as I say, it's a pity.

These are my files: Any help with this problem was greatly appreciated