DFB1r4 design discussion thread

General discussions or ideas about hardware.
User avatar
Badwolf
Posts: 2228
Joined: Tue Nov 19, 2019 12:09 pm

Re: DFB1r4 design discussion thread

Post by Badwolf »

exxos wrote: Tue Sep 28, 2021 12:40 am I assume you isolating AS30 from access zero and you decoding it yourself ? Only thing I can think of currently. As if the ST decodes zero then it will be based on 8mhz clocks and keep DTACK low for A while..
My external ROM is actually disabled throughout this (didn't want to introduce additional issues), so address zero, plus all the normal Exxxxx space is being decoded and serviced by the motherboard.

I'm back to my old clock switching code so the clock is only ever at the motherboard's speed when asserting XAS. Quite, therefore, why there's a problem around reset, I've not worked out -- but holding off the 'allow acceleration' flag until later in the boot process seems to work.

In effect, it's not doing any clock switching until the OS tries to make the jump to 16MHz (I'm decoding the call to that bus control register).

Odd, but not sure I'm going to go chasing why just yet.

BW
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
DSTB1 Open source 16Mhz 68k and AltRAM accelerator for the ST
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
User avatar
Badwolf
Posts: 2228
Joined: Tue Nov 19, 2019 12:09 pm

Re: DFB1r4 design discussion thread

Post by Badwolf »

So I've been doing some work on the SDRAM recently.

My soak testing has been going well: two hours of Doom first, then YAARTTT under MiNT. No errors reported:

IMG_4783.jpeg
IMG_4783.jpeg (191.82 KiB) Viewed 1872 times

Then the minute I shut down the Toswin2 window, I get an odd freeze where the mouse still moves, but nothing is clickable and keyboard doesn't respond, even to CTRL-ALT-DEL. I can see increased ROM access when moving the mouse. Bizarre and frustrating.

Anyway, I can't stare at RAM logic any more, so thought I'd return to the FPU. Enabled it in the firmware and popped in my new chip: hang on boot.

Odd, try it at 16MHz. Same. Try in TOS: this boots, but 13 bombs when accessing the chip. Crap. I hope my new chip's not toast.

So I remove the board and put the chip into the Falcon motherboard. Boot up. No FPU picked up. Damn and blast. Looks like I've somehow blown another chip.

So decide to put my 'partially' working one that I thought I'd blown up before (it's recognised, you can measure its speed, but all the results are wrong). It's perfect! Passing the FPUTEST program...

These bloody PLCC sockets must make *terrible* connections on occasion. What else could it be? I'm going to go and clean up my chip a bit more, but hell's teeth, what a palaver. I'm very tempted to get a PGA FPU next time, albeit I have no idea if the sockets are available.

BW
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
DSTB1 Open source 16Mhz 68k and AltRAM accelerator for the ST
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
Steve
Posts: 2570
Joined: Fri Sep 15, 2017 11:49 am

Re: DFB1r4 design discussion thread

Post by Steve »

@Badwolf yeah they are really really crappy (the plcc sockets) The amount of times I've had to go and individually get a needle and pull the teeth out a bit to make better contacts. Also look at the bottom of the FPU chip, the legs curve and sit inside a little recess on the bottom, I usually get a tiny flat screwdriver and push the legs towards the outside edge of the recess to make the legs more prominent. Fixed many an issue like this.
User avatar
Badwolf
Posts: 2228
Joined: Tue Nov 19, 2019 12:09 pm

Re: DFB1r4 design discussion thread

Post by Badwolf »

Steve wrote: Thu Oct 07, 2021 11:48 am @Badwolf yeah they are really really crappy (the plcc sockets) The amount of times I've had to go and individually get a needle and pull the teeth out a bit to make better contacts. Also look at the bottom of the FPU chip, the legs curve and sit inside a little recess on the bottom, I usually get a tiny flat screwdriver and push the legs towards the outside edge of the recess to make the legs more prominent. Fixed many an issue like this.
I made a last minute switch from SMT to through-hole sockets (as then I don't have to cut the base off and stick it back on afterwards -- I thought it would be an upgrade). This is increasingly looking like a bad idea because not only are these through-hole pins genuinely flimsier than a poundshop brolly in a stiff breeze, I can't now just say 'Donald Duck it' and solder the chip directly onto the pads!

We really need that 030-but-no-FPU GCC toolchain. ;)

BW
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
DSTB1 Open source 16Mhz 68k and AltRAM accelerator for the ST
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
User avatar
exxos
Site Admin
Site Admin
Posts: 23488
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: DFB1r4 design discussion thread

Post by exxos »

Badwolf wrote: Thu Oct 07, 2021 3:17 pm This is increasingly looking like a bad idea because not only are these through-hole pins genuinely flimsier than a poundshop brolly in a stiff breeze

I had that problem with those "tru-connect" DIP sockets from rapid. I documented it somewhere on my H5 build. That I literally just pushed the chip in removed it pushed it back in and it broke some of the pins on the socket! upon closer inspection those sockets seem to use tinfoil for the pins they are that thin and flimsy. I even mentioned this to a rapid sales rep when they were pestering me a while ago and did not seem bothered in the slightest. Annoying as I basically have a otherwise good H5 somewhere just with loads of cheap sockets which are all broken now :roll:

I mean yeah, the 3M sockets are like 5x the price of everything else, but how long would it take me to remove and refit all the sockets on the motherboard..It is actually one of the reasons why the H5 price got ramped a bit, as I think some people use the cheap sockets, then like the 3M are like £3 each or something.. I'm just done screwing around with any other brands of sockets these days.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
User avatar
mrbombermillzy
Posts: 1441
Joined: Sun Jun 03, 2018 7:37 pm

Re: DFB1r4 design discussion thread

Post by mrbombermillzy »

exxos wrote: Thu Oct 07, 2021 3:26 pm I mean yeah, the 3M sockets are like 5x the price of everything else, but how long would it take me to remove and refit all the sockets on the motherboard..It is actually one of the reasons why the H5 price got ramped a bit, as I think some people use the cheap sockets, then like the 3M are like £3 each or something.. I'm just done screwing around with any other brands of sockets these days.
Here here...
They should add an extra pig to the famous childrens fable, that used cheap Chinese 'home made' steel/electronic components to build his house...

You get what you pay for; literally.
User avatar
Badwolf
Posts: 2228
Joined: Tue Nov 19, 2019 12:09 pm

Re: DFB1r4 design discussion thread

Post by Badwolf »

Many thanks to the Exxos store fairy who has delivered a pair of 3M PLCC68 sockets (and a most welcome bonus too!)

You're right. These don't exactly have girders protruding from the bottom of the socket, but they're noticably stronger than whatever the heck things I've put on my board. This is despite the mould being absolutely identical.

Not all sockets are created equal!

BW
DFB1 Open source 50MHz 030 and TT-RAM accelerator for the Falcon
DSTB1 Open source 16Mhz 68k and AltRAM accelerator for the ST
Smalliermouse ST-optimised USB mouse adapter based on SmallyMouse2
FrontBench The Frontier: Elite 2 intro as a benchmark
User avatar
exxos
Site Admin
Site Admin
Posts: 23488
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: DFB1r4 design discussion thread

Post by exxos »

Yeah there is room for improvement on the 3M sockets. I wish the pins were wider. But I've not had any break.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
User avatar
JezC
Posts: 2081
Joined: Mon Aug 28, 2017 11:44 pm

Re: DFB1r4 design discussion thread

Post by JezC »

Another annoying socket-related issue is that Farnell seem to have stopped offering 3M sockets on at least some PLCC sizes, so I'm concerned about where I can get some reliable 84 pin PLCC sockets for my two IDE v1.1C boards...and many others are shown as available until stocks are exhausted... :(
User avatar
exxos
Site Admin
Site Admin
Posts: 23488
Joined: Wed Aug 16, 2017 11:19 pm
Location: UK
Contact:

Re: DFB1r4 design discussion thread

Post by exxos »

I found the same with farnell :( but I will have them in my store at some point anyway. I bulk buy things these days. I've got the 84pins in my store, but they are not 3M but don't think I had any problems with them.
https://www.exxosforum.co.uk/atari/ All my hardware guides - mods - games - STOS
https://www.exxosforum.co.uk/atari/store2/ - All my hardware mods for sale - Please help support by making a purchase.
viewtopic.php?f=17&t=1585 Have you done the Mandatory Fixes ?
Just because a lot of people agree on something, doesn't make it a fact. ~exxos ~
People should find solutions to problems, not find problems with solutions.
Post Reply

Return to “HARDWARE DISCUSSIONS”