Hypothetically, what could be gained from a developer side if SRAM was supported?

For topics which do not fit in other specific forums.
User avatar
SegaSnatcher
Posts: 163
Joined: Sun May 24, 2020 9:18 pm
Has thanked: 36 times
Been thanked: 43 times

Hypothetically, what could be gained from a developer side if SRAM was supported?

Unread post by SegaSnatcher »

Just a thought experiment and obviously would have to be limited to Digital IO boards with the extra 40 GPIO pins, but let's say just for fun an add on SRAM memory board was available to developers, what gains could they possibly see while developing cores? I hear SRAM in terms of performance is 2nd after BRAM. Not saying an SRAM board would ever be implemented or needed, but I would assume in some situation it could possibly be a benefit.
User avatar
Newsdee
Top Contributor
Posts: 830
Joined: Mon May 25, 2020 1:07 am
Has thanked: 98 times
Been thanked: 209 times

Re: Hypothetically, what could be gained from a developer side if SRAM was supported?

Unread post by Newsdee »

From what I know, having an additional fast RAM bank (even a small one) can make writing/porting some cores easier, but it's not guaranteed.

There's some ZX-Uno cores written for 2MB of SRAM which I think are difficult to port due to RAM access timings they rely on; but they're not much to be excited about considering other cores available on MiSTer (e.g. the PC-186: https://www.zxuno.com/forum/viewtopic.php?t=815 )

If somebody were to write a new core, having more available RAM banks makes things a little easier just because you don't have to come up with (simple) tricks to handle memory access without clashes. But that doesn't mean they can't be solved, Sorg and others did some magic with Neogeo and ao486 already...
Post Reply