Shponglefan wrote on Yesterday, 12:06:

dormcat wrote on Yesterday, 08:22:

Shponglefan wrote on Yesterday, 01:15:

For the ISA card, a switch is hooked up to what I think is the bus timing jumper. With the jumper closed, the ISA card is enabled and boots. With the jumper open, the ISA card won't boot. Then either the AGP or PCI card will function depending on the BIOS initialization priority setting.

So the ISA card can override AGP/PCI cards with just a switch? That's something new to know; good work! 👍

Normally ISA video will always take priority. In this case removing the bus timing jumper seems to prevent the card's video bios from loading, so it's effectively disabled.

And I can't take full credit for this, I came across the idea in an 11-year old thread here: 486 Mobos: Can you install multiple VGA cards - like AGP/ PCI switching?

Disabling the video bios sounds like a more universal approach?

Something that just disables the address pins of the chip if disabled?

A latch chip / bus transciever tucked in 'dead bug' style inside a socket, with jumper pushed out one side, that goes in the vbios socket, with the vbios on top, perhaps?

When jumper is on, latch is on, and vbios address pins can be raised. When jumper is off, latch is off, and address pins cannot be raised.

Parasitically uses sockect's vcc and gnd to power the latch?

If the vbios uses CSelect, then we could just use a single transistor? (With the jumper on the 'on' signal for the transistor, with CSel signal on the collector, CSel status would go right through,and the vbios would behave normally. With the jumper removed, the transistor is 'off', and it wont.)

I finally had half an hour in the lab to test out the monitoring prototype on my "unloved" 9500Pro - And the results look good already.
After both my MCP2221 and CP2112 annoyed me because of reasons ("hurdur I don't wanna talk i2c today") I just slapped on a raspberry Pi zero W with a sole purpose: I2C-tools.

Interestingly I had booted up that pi last 1 year ago.. and apparently for the same purpose.. 😁 - hostname was "iic".

After having a short laugh, I had some tinkering on the firmware and threw a first validation python script into the mix to see, if the results are plausible.

And yes, they are!

These values are after around 7 minutes of S.T.A.L.K.E.R. - Shadow of Chernobyl. More tweaking etc to come, but the basic validation looks promising.

tehsiggi wrote on Yesterday, 17:28:

And yes, they are!

The attachment 5min_stalker_shoc.png is no longer available

These values are after around 7 minutes of S.T.A.L.K.E.R. - Shadow of Chernobyl. More tweaking etc to come, but the basic validation looks promising.

We need thumbs up buttons . This is cool.

I started playing with my vintage serial pen plotter again.

Fancy experiments with novel medium inside a washed and refilled Pilot V5 precise pen.

It looks like Rit DyeMore dye mixed with propylene gycol, will work here. More experiments needed.

Possibly more permanent / lightfast than the inkjet ink I've been using.

Long duration plot needs a few more DAYS to complete. Then it's into the window to learn more.

Checked out Neocities and began cataloguing some of my PCI sound cards in delicious HTML.

But, I've used PNG and maybe I should have used retrosafe GIF instead?

https://mattrocks.neocities.org

PNG for photos is a bit of a waste of bandwidth... JPG will be a lot better choice, and also on older browsers... if they could access the https at all 🤣

That's a good point. HTTPS defeats all forms of graceful degradation, and I think everyone has broadband too. At least using uncompressed PNG restores the slow dial-up experience we all had 😉

I managed to get my (family's) old copy of Lemmings for Win95 running today on my main daily-driver PC (running fedora 43 kde), after a... few hours wasted trying to get the original Windows MIDI soundfont to work before realising I was messing with the wrong config file.

I've been thinking about some possibly misplaced memories. I remember seeing the "It's now safe to turn off your computer" prompt on my first PC, but after a lot of searching I'm sure it was from a series of 430TX PCs that only have ATX power. And there's no option in the fairly locked down BIOS to disable ACPI. Is there something that could explain this?

giantenemycat wrote on Yesterday, 23:27:

I've been thinking about some possibly misplaced memories. I remember seeing the "It's now safe to turn off your computer" prompt on my first PC, but after a lot of searching I'm sure it was from a series of 430TX PCs that only have ATX power. And there's no option in the fairly locked down BIOS to disable ACPI. Is there something that could explain this?

It was typical for suppliers to install Windows on default BIOS settings. And, I think the BIOS needs to be reconfigured before Windows is installed otherwise you could have ATX APM and the shutdown screen.

Also, I have never seen an i430TX in ATX form factor with soft power button on the case.

Tested out a P3 900 (100MHZ FSB version) that I got today and it works nicely at stock speed but unfortunately does not post at 1200 (9X133).

Ah well it was worth a try.

The Sertek ES688 and ES1688 sound cards share the same PCB design; the only difference lies in their ESS chips and OPL3 components—the ES688 features a Yamaha OPL3, while the ES1688 utilizes ESFM. A common characteristic of both cards is that their Line Out output circuits contain only a single 100nF MLCC capacitor, resulting in a noticeably thin bass response. To address this, I removed the original capacitor and installed a 22µF tantalum capacitor in its place, while also adding a 100nF MLCC capacitor in parallel across the tantalum capacitor. Following this modification, the bass feels significantly more powerful and robust.

zwrr wrote on Today, 06:11:

The Sertek ES688 and ES1688 sound cards share the same PCB design; the only difference lies in their ESS chips and OPL3 components—the ES688 features a Yamaha OPL3, while the ES1688 utilizes ESFM. A common characteristic of both cards is that their Line Out output circuits contain only a single 100nF MLCC capacitor, resulting in a noticeably thin bass response. To address this, I removed the original capacitor and installed a 22µF tantalum capacitor in its place, while also adding a 100nF MLCC capacitor in parallel across the tantalum capacitor. Following this modification, the bass feels significantly more powerful and robust.

The attachment ES688.jpg is no longer available

We're talking DC-decoupling caps here? Like in series with the signal? 100nF ceramics default are - well, not good - for two reasons: frequency cutoff because of capacitance and their piezoelectric properties.

Tantalum have higher capacity.. however they're also prone to piezoelectric effects and iirc have bad signal influence in AC (dirstortion) when lowly biased. I guess everything is better than 100nF, but you could just have kept the tantals alone, the ceramics with their 100nF are negligible.

"better" would have been a simple electrolytic or even better (but space is an issue) film capacitor.

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now.

The RAM + CPU board looks like this (still old RAM sticks here)

And after the upgrade. Yes, the SSD is almost smaller.

Minutemanqvs wrote on Today, 10:58:

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now. […]
Show full quote

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now.

The RAM + CPU board looks like this (still old RAM sticks here)

And after the upgrade. Yes, the SSD is almost smaller.

The next incoming upgrade are 2 Xeon X5680 which will give me 12 cores and a 1333MHz memory speed.

Minutemanqvs wrote on Today, 10:58:

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now. […]
Show full quote

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now.

The RAM + CPU board looks like this (still old RAM sticks here)

And after the upgrade. Yes, the SSD is almost smaller.

That's always nice to see 😀

Minutemanqvs wrote on Today, 10:59:

Minutemanqvs wrote on Today, 10:58:

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now. […]
Show full quote

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now.

The RAM + CPU board looks like this (still old RAM sticks here)

And after the upgrade. Yes, the SSD is almost smaller.

The next incoming upgrade are 2 Xeon X5680 which will give me 12 cores and a 1333MHz memory speed.

Screams RAM disk 😉

tehsiggi wrote on Today, 11:07:

Minutemanqvs wrote on Today, 10:59:

Minutemanqvs wrote on Today, 10:58:

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now. […]
Show full quote

I already did this a couple of days ago, but I upgraded my 2010 Mac Pro 5,1 to 128GB RAM, just because RAM is cheap now.

The RAM + CPU board looks like this (still old RAM sticks here)

And after the upgrade. Yes, the SSD is almost smaller.

The next incoming upgrade are 2 Xeon X5680 which will give me 12 cores and a 1333MHz memory speed.

Screams RAM disk 😉

I'm in search of a problem to solve with all that RAM 😁

I treated myself to several Flash Carts, which all arrived over the past couple of days, so I've been busy testing them all out.

I got an A8PicoCart for my Atari 65XE computer, enabling me to play cartridge ROMs. A nice accompaniment for the SDrive-Max that I already have, allowing me to play the whole game library. This is working well, didn't really need any tinkering.

Also got a Saroo from China, for the Sega Saturn. I should have gotten this sooner, the convenience of not having to swap out discs is great! I've tested a bunch of random games and everything I've tried so far has worked just fine.

Finally, also from China, ordered an SD2SNES cartridge for my Super Famicom. This arrived with a broken microSD card slot, so I had to desolder that and solder in a replacement. It all went well, considering this was my first time working with surface mount soldering and super tiny pins! I was pleased and relieved to see it finally boot, allowing me to test it out. I've mostly been playing around with MSU-1 ROM hacks, which are quite cool.

Did end the chapter on the Connect C3D-6055 Radeon 9550 card that was claiming universal support. It doesn't - the A2 pin is wired to a zero ohm resistor.

Stuck it into a fresh Soltek QBIC E3702M since that one uses nForce 2. For those curious, the QBIC is sort of a Shuttle XPC "clone", except it doesn't have the wallmounted cooler that XPC uses, rather you can use your own heatsink. (as long as it doesn't interfere with the ATX harness.)

Also PSU is SFX which makes it infinitely easier to source over Shuttle's FlexATX format PSU.