Good to know. Please let me know if there is any important information, but as soon as my Syncro030/S is shipped out and arrives to me and I will try it out!
Synchr030/S 256MB SDRAM Accelerator for the SE/30
- Thread starter YMK
- Start date
I already had a sidebar going with this user predating this thread, but I will lay it out here in more detail. This is uncharted territory, I do not have a Synchr030 to test with and only performed some testing with the booster while I had one on loan. I don't recall if I tried Synchr030 with a 30Video card or any other combination thereof.
There's a few other things to unpack here, so I'll clarify in no particular order.
In general, any more than 1 card in a SE/30 (whether it's located in PDS or CPU socket) is not a guaranteed to work scenario. Most combinations of two cards are fine, but 3 card configurations are a wildcard. The only 3 card configuration I'll say definitely works is Bolle Riser + Carrera/Powercache/Turbo040 + 30Video (any model).
There are a few requirements of any card with a passthrough slot in order to improve compatibility with additional cards and accelerators. One of those is buffering of the system clock for downstream cards as the SE/30 system clock is weak to begin with and with multiple cards it degrades badly. I do this on all of my boards with passthrough slots that use the system clock signal.
Booster configurations: Accelerators should be located as close to the logic board as possible. In this case that's definitely the preferred option here as that arrangement will allow the booster to buffer the clock for the Synchro and any other cards downstream. I don't believe that clock buffering is done on the Synchr030, so that's part of the reason I think that strong pull up resistor on the clock can allow the booster on top configuration to work. From what I found all clocks were stable on my board and had the expected phase relationships, so the exact mechanism of action with the resistor is unclear here (however I don't know what requirements/tolerances the Synchr030 has).
My stance on the clock pull up: If the resistor works for you, great, it's probably fine but it's not something I "bless" as a supported way to make that configuration always work especially as why it helps is not known.
30Video cards are not clones, they are a scratch designed board and driver I created last year. The HC and SI cards do not depend on or use the system clock and should be immune from any issues with it. The GS card does use the system clock, if no other clock in the system buffers the clock signal the GS card should be first in the stack to allow it to do so. Anything but that is an untested/non-recommended configuration. Not to say it can't work, but if any issues occur/system instability (well, beyond the usual for Mac OS) that's a possible root cause.
The 30Video cards are very compatible (I've never run into a scenario where they don't co-exist with another board of any type) so I would not expect an issue with Synchr030 itself, keeping in mind the 1-card-per-system intended spec for these systems. The 30Video cards are all fairly closely related so where one works, I would expect the other models to also work. I am aware that it's been tested with the Synchr030 at least a couple of times in the GS version as seen above. I don't know if anyone is running a finished setup with a Synchr30+30Video though.
There's a few other things to unpack here, so I'll clarify in no particular order.
In general, any more than 1 card in a SE/30 (whether it's located in PDS or CPU socket) is not a guaranteed to work scenario. Most combinations of two cards are fine, but 3 card configurations are a wildcard. The only 3 card configuration I'll say definitely works is Bolle Riser + Carrera/Powercache/Turbo040 + 30Video (any model).
There are a few requirements of any card with a passthrough slot in order to improve compatibility with additional cards and accelerators. One of those is buffering of the system clock for downstream cards as the SE/30 system clock is weak to begin with and with multiple cards it degrades badly. I do this on all of my boards with passthrough slots that use the system clock signal.
Booster configurations: Accelerators should be located as close to the logic board as possible. In this case that's definitely the preferred option here as that arrangement will allow the booster to buffer the clock for the Synchro and any other cards downstream. I don't believe that clock buffering is done on the Synchr030, so that's part of the reason I think that strong pull up resistor on the clock can allow the booster on top configuration to work. From what I found all clocks were stable on my board and had the expected phase relationships, so the exact mechanism of action with the resistor is unclear here (however I don't know what requirements/tolerances the Synchr030 has).
My stance on the clock pull up: If the resistor works for you, great, it's probably fine but it's not something I "bless" as a supported way to make that configuration always work especially as why it helps is not known.
30Video cards are not clones, they are a scratch designed board and driver I created last year. The HC and SI cards do not depend on or use the system clock and should be immune from any issues with it. The GS card does use the system clock, if no other clock in the system buffers the clock signal the GS card should be first in the stack to allow it to do so. Anything but that is an untested/non-recommended configuration. Not to say it can't work, but if any issues occur/system instability (well, beyond the usual for Mac OS) that's a possible root cause.
The 30Video cards are very compatible (I've never run into a scenario where they don't co-exist with another board of any type) so I would not expect an issue with Synchr030 itself, keeping in mind the 1-card-per-system intended spec for these systems. The 30Video cards are all fairly closely related so where one works, I would expect the other models to also work. I am aware that it's been tested with the Synchr030 at least a couple of times in the GS version as seen above. I don't know if anyone is running a finished setup with a Synchr30+30Video though.
Yes, I forgot to mention that I had talked to you. I'm so sorry about that!
Thanks for the in-detail explanation, that is very good to know. I will let everyone know if it works and if not, what issues I am having. Thank you so much to everyone who is contributing over here, I can't stress enough how cool this forum is!
Thanks for the in-detail explanation, that is very good to know. I will let everyone know if it works and if not, what issues I am having. Thank you so much to everyone who is contributing over here, I can't stress enough how cool this forum is!
This is all incredible. I’ve been away from the vintage Mac thing for a while, but am refurbishing my garage.
@YMK does the Synchr030/S work with Optima under System 6?
@YMK does the Synchr030/S work with Optima under System 6?
I never tested Optima, so I have no idea.
I searched the Synchr030/S documentation, but there is no mention of even System 6 or 24-bit addressing.
I searched through past emails between myself and YMK (because there are so many), and I couldn't find "Optima" as a keyword. I then searched for 24-bit addressing in our email dialog and found one email from January 2025 saying:
Of course, the above statement did not take into consideration Optima's tricks of implementing 32-bit addressing (and therefore, access to more than 8MB of RAM) under System 6. So it boils down to the question of whether the Synchr030/S Extension is compatible with the Optima INITs, and again, I don't know that because I never tested it.
What I CAN say is that you get 128MB of RAM when the Extension is NOT loaded (not the full 256MB), which is the case running System 6 normally, without Optima installed. You are simply limited to 8MB of usable RAM (with 24-bit addressing), with the remaining 120MB being reserved for you to use as a RAM disk.
I haven't tried this. Let me get back to you.
Edit: Is there any reason to use Optima over Maxima?
Last edited:
The Synchr030/S documentation mentions a non-stock ROM as being very important, and in my video I stated it as being pretty much a requirement...
"Required"? Really?
Well yes, and for more reasons than just to become 32-bit clean. Because unless you want to sit through RAM checking for 384MB (it would check 128MB, then the Extension would load, trigger a reboot, and then it would check the full 256MB), a non-stock ROM is essential to practical use of Synchr030/S. That's because non-stock ROMs kill the infuriatingly slow RAM checks at cold boot. Honestly, I wish Apple had never implemented that checking. It's so incredibly bothersome!
The good news is there are many ROM SIMMs to choose from, as I mention here.
Also note that simply having a ROM-inator II (or similar) to make your machine 32-bit clean doesn't magically make System 6 32-bit clean. System 6 (without Optima) is 24-bit only, regardless of your ROM SIMM. Optima promises to change that, but I've read about many software incompatibilities. So even if Synchr030/S is found to be compatible with Optima, one would need to contend with those incompatibilities.
Agreed. The RAM check would have been better triggered by a certain startup key if RAM is suspect.
Enabling the full 256MB depends on certain code being present in a IIsi derived ROM, not just for the practical reason of skipping RAM tests. In other words, you will not get 256MB with the stock ROM.