REMINDER - Stay logged in for at least 2 hours a week to get whitelisted.
Also it helps build a picture where our "good traffic" is coming from for detection scripts.
:o)

Booster clock switching + DMA nerding

Discussion and support for the DSTB1 & DFB1 boosters by BadWolf..
User avatar
JezC
Posts: 2789
Joined: 28 Aug 2017 23:44

Re: Booster clock switching + DMA nerding

Post by JezC »

Badwolf wrote: 26 Jul 2024 16:26
JezC wrote: 26 Jul 2024 11:29 @Badwolf I'll be interested to know if/when you think you have a final solution/version to test...
I think I do now, but it requires some physical modifications which I need to document. A new revision is probably on the cards.

If you're happy to cut a few traces and wire in a few bodges, then I'd be keen to get additional feedback.

BW
Yes, I'll probably finish the original build & briefly test it & then do the mods once know it's basically working.

No rush, just update us when you can! Thanks.
User avatar
Badwolf
Site sponsor
Site sponsor
Posts: 3048
Joined: 19 Nov 2019 12:09

Re: Booster clock switching + DMA nerding

Post by Badwolf »

JezC wrote: 26 Jul 2024 21:45
Badwolf wrote: 26 Jul 2024 16:26

I think I do now, but it requires some physical modifications which I need to document. A new revision is probably on the cards.

If you're happy to cut a few traces and wire in a few bodges, then I'd be keen to get additional feedback.

BW
Yes, I'll probably finish the original build & briefly test it & then do the mods once know it's basically working.

No rush, just update us when you can! Thanks.
If you've not yet put on the socket strip and the pin headers, it's easier to cut the traces.

Basically you need to:

* Cut BG to BG on the headers (tiny trace between DIP64 pins 11.
* Cut the VCC line going to the CKE pin on the SDRAM (the short 45 degree trace to a via coming out of pin 37 on the RAM chip)
* Run a bodge wire from the BG line going to the motherboard to TP4
* Run a bodge wire from the BG line going to the CPU to TP5
* Run a bodge wire to the CKE pin on the SDRAM chip to TP3

Here's how I did it:
IMG_8002.jpeg
IMG_8003.jpeg

TP1 remains the TOS 2 decoder.
I've a plan for TP2, but probably only @coonsgm would be interested in that...

BW
You do not have the required permissions to view the files attached to this post.
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
coonsgm
Posts: 454
Joined: 30 Jan 2021 01:30

Re: Booster clock switching + DMA nerding

Post by coonsgm »

What is this I would be interested in??? TP2 you say...
User avatar
agranlund
Site sponsor
Site sponsor
Posts: 1756
Joined: 18 Aug 2019 22:43
Location: Sweden

Re: Booster clock switching + DMA nerding

Post by agranlund »

coonsgm wrote: 27 Jul 2024 16:18 What is this I would be interested in??? TP2 you say...
TurboPower2 ? :)
coonsgm
Posts: 454
Joined: 30 Jan 2021 01:30

Re: Booster clock switching + DMA nerding

Post by coonsgm »

agranlund wrote: 28 Jul 2024 14:13
coonsgm wrote: 27 Jul 2024 16:18 What is this I would be interested in??? TP2 you say...
TurboPower2 ? :)
Well then, I am 100% for more power. Turbopower is just that much better.
User avatar
JezC
Posts: 2789
Joined: 28 Aug 2017 23:44

Re: Booster clock switching + DMA nerding

Post by JezC »

@Badwolf I'll dig out the board & check if I'd soldered the socket/headers on or not...been so long I honestly can't remember!
:roll: :?
User avatar
Badwolf
Site sponsor
Site sponsor
Posts: 3048
Joined: 19 Nov 2019 12:09

Re: Booster clock switching + DMA nerding

Post by Badwolf »

coonsgm wrote: 28 Jul 2024 19:07
agranlund wrote: 28 Jul 2024 14:13 TurboPower2 ? :)
Well then, I am 100% for more power. Turbopower is just that much better.
Haha.

No, TP2 is Test Point 2. DSTB has five test points and four have now been co-opted by other functions, but TP2 appeared available.

I had it in my mind that you talked about measuring the effectiveness of accelerators on games in one of your videos -- or perhaps it was in a forum post. Anyway, one of my old dev boards pulsed a line when a write was made to the video shifter registers. Implication being that a double-buffering switch was occurring. I thought I may add that to TP2 on DSTB1 by default.

With a suitable frequency counter, it lets you measure the instantaneous FPS of any double-buffered game. Just set up a vaguely stable and repeatable section of play and note the frame rate. Try again with the booster.

8MHz STRAM:
IMG_8015.jpeg

16MHz ALTRAM:
IMG_8014.jpeg


8MHz STRAM:
IMG_8012.jpeg

16MHz STRAM (game doesn't support running in AltRAM):
IMG_8011.jpeg

We can therefore conclude Stunt Car Racer isn't really CPU limited and is memory limited instead. It would need a patch to let AltRAM kick in, or a MSTE-style cache to get much boost.

Frontier, OTOH, because it supports AltRAM (by accident -- hard coded frame buffer addresses), gains a healthy boost.

BW
You do not have the required permissions to view the files attached to this post.
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
coonsgm
Posts: 454
Joined: 30 Jan 2021 01:30

Re: Booster clock switching + DMA nerding

Post by coonsgm »

This is an awesome way to technically measure the performance goodness and better understand where the bottleneck is in the system.

Return to “DSTB1 & DFB1 booster by BadWolf”

Who is online

Users browsing this forum: ClaudeBot and 1 guest