Quadra 610 PDS slot vs. an "LC PDS" slot / Apple IIe card?

[Realitystorm]

I was wondering, since the Quadra 610 supports 24-bit mode, and has a PDS slot, will it support a Apple IIe card?  It's not on the official list, so I think the answer is likely no, but thought I'd put it out there in case someone has tried and it works.

I dodn't know what the physical difference was between an LC PDS vs. a regular PDS, I assumed an LC PDS is horizontal mount vs. vertical.  The Quadra 610 looks like it has a horizontal mount too though.

 

[beachycove]

There were a number of PDS slots, and the two in this case were not mechanically or electronically compatible. There was a (rare, Rare, RARE) Nubus adapter that could be used in the Q610 PDS slot, but no 680030-PDS-to-Q610-PDS adapter.

 

[Realitystorm]

There were a number of PDS slots, and the two in this case were not mechanically or electronically compatible. There was a (rare, Rare, RARE) Nubus adapter that could be used in the Q610 PDS slot, but no 680030-PDS-to-Q610-PDS adapter.

Thanks, I read up a bit more after posting and found out the same, it had to be specifically a LC PDS slot for compatibility, before that they were to closely tied to the specific CPU / Bus speed.

 

[Franklinstein]

No, the Q/C 610 (or later 660av) cannot support the Apple IIe card, primarily because this computer uses either a 68040 PDS or NuBus (if you have the proper adapter, as noted above). The most common thing plugged into a 610's expansion slot was the 66MHz 486-based PC Compatibility Card, which ran off of the 68040 PDS in a right-angle adapter.

The Apple IIe card was specifically designed for the Macintosh LC and its 16-bit, 16MHz LC PDS. It will run in many other computers that have an LC PDS but requires the host computer be able to support 24-bit addressing. I believe the LC 475 and LC 575 are able to use it but I'm not sure about the LC 630 or other IDE-based 040 Macs. None of the Power Macs can use it.

As noted, there were many Processor Direct Slots, and most of them were not interchangeable. There are the machine-specific slots for the SE, SE/30, IIsi, IIci, IIcx, IIvi, IIvx, and a handful of derivative models. The multi-model PDSes were used in basically three machines: the 68040 PDS in the high-end Centris/Quadras, the 601 PDS in the NuBus Power Macs, and the LC/LC III PDS. I think the original Comm Slot was configured as some sort of psuedo PDS/serial port pass-through but I'm not sure; it was only used on 68040 boxes with slide-out logic boards like the 630 and 575. If used, it was usually filled with a modem; the only other option was an Ethernet card. Same with the CS II, only it was PCI-based so it was much faster if used for Ethernet.

The LC PDS is the only widely used multi-generational PDS. Introduced with the Macintosh LC (hence the name), it was originally a 16-bit interface connected directly to the bus of the 68020. When the LC II was introduced, it kept the PDS with no changes, since none were necessary. The Color Classic also adopted the LC PDS slot for expansion. The LC III was introduced with essentially the same slot but with an extra set of pins on the back, which when used provided a full 32-bit path to the processor, but otherwise was completely compatible with the original LC PDS. The LC 520 and Color Classic II also adopted the LC III PDS, and it was still directly connected to the processor bus. 

This changed with the introduction of the 68040: the new processor had a different bus, so nothing that ran on the 030 bus would run properly on the 040 bus. However, Apple wanted to keep compatibility for older devices on the consumer/education-level machines, so they built a custom chip that emulated the 030 bus and interfaced with both the LC PDS and new Comm Slot devices.

The introduction of PowerPC didn't change much of anything: Apple simply tweaked the 030 bus emulation chip and connected it to a new chip that bridged PowerPC to an 040 bus (which is why the 52/62/53/63xx machines are so slow: there are three different buses talking through two bridge chips).

 

[Trash80toHP_Mini]

Great summary of Apple's infernal incompatible PDS implementations. Thought that last bit's not quite accurate as I understand it.  @Cory5412 posted a link to an excellent article that thoroughly debunked lowendmac's error ridden RoadApple PPC myth. Only one device can talk to the processor at a time and anything on the downstream side of the bridge had a much lower bandwidth requirement than the CPU bus, so the bridge imposed no real penalty in those machines.

Bridged 030 "slow I/O bus" implementations worked well for the Mac all the way up through the PBX memory controller/bridge ASIC in the PowerBook 1400. Intel finally opened its closely held Front Side Bus equivalent to outside developers in 1997, about the same time the 1400 with its PBX bridged '030 I/O bus was discontinued.

 

[Gorgonops]

Intel finally opened its closely held Front Side Bus equivalent to outside developers in 1997

Uhm, do you have a citation as to what you're referencing here? I don't think Intel demanded that third-parties pay licensing fees to develop bus chipsets until the Pentium 4. (They slapped a lot of patents on the P4's quad-pumped bus architecture.) That was well *after* 1997.

 

[Trash80toHP_Mini]

HRMMM? Dunno if my understanding of it is correct, doesn't sound so much like licensing as outright denial of use. It's in Wikipedia's Front Side Bus article.

 

[Gorgonops]

.... Only one device can talk to the processor at a time and anything on the downstream side of the bridge had a much lower bandwidth requirement than the CPU bus, so the bridge imposed no real penalty in those machines.

Bridged 030 "slow I/O bus" implementations worked well for the Mac all the way up through the PBX memory controller/bridge ASIC in the PowerBook 1400...

To put it another way, the LC "030" bus sort of became the low-end Mac equivalent of the function 16-bit ISA served on PCs well into the early 2000s. Generally speaking there's nothing terribly wrong with hanging a device off a slow bus as long as the bandwidth requirements of that device don't exceed the capacity of said bus.

Generally speaking the real reason the 52/62/53/63xx sucked so badly is the original 603 had a smaller split cache (8k each instructions/data vs. 32k unified) than the 601 and the version of the 68k emulation software built into their ROMs didn't work anywhere near as well with it. Considering almost the entirety of the OS ran under emulation in System 7 that was a big problem. This is why those machines show significant improvement when running under OS 8 and later. (A lot more of the OS becomes native.)

 

[Gorgonops]

HRMMM? Dunno if my understanding of it is correct, doesn't sound so much like licensing as outright denial of use. It's in Wikipedia's Front Side Bus article.

So, the date in that article is 2007, not 1997, and what they're talking about is allowing non-Intel devices to reside in the Pentium 4/Core CPU socket. (IE, remember the aforementioned licensing of the patent-encumbered Pentium 4 bus? Intel granted some licenses to make Pentium 4 *chipsets*, but prior to this deal they'd never granted anyone a license to make a "CPU" that fit it. That's why some also-ran companies like VIA were continuing to make CPUs that fit the Pentium 3 socket for long after it wasn't really viable anymore.)

That article is frankly pretty terrible. That criticism applies generally to it but that paragraph in particular makes no sense without putting it into the context of Intel's Pentium 4 bus licensing scheme.

 

[Trash80toHP_Mini]

OOPSIE, my bad, not enough caffeine coursing through the brainpan yet apparently.  [:I] Thanks for the correction.

Thanks also for putting it a better way regarding the bridged bus bit. Didn't know about the OS7/OS8 performance divide for that series, I'll have to check which was on the 6290 I was gifted. Knowing the giver, I'll bet it was some flavor of OS8 which would explain much about why I never thought of it as particularly slow whenever I used it in the rug rat's room.

Found the bookmark: http://www.taylordesign.net/classic-macintosh/the-mythical-road-apple/

 

[Franklinstein]

Great summary of Apple's infernal incompatible PDS implementations. Thought that last bit's not quite accurate as I understand it.  @Cory5412 posted a link to an excellent article that thoroughly debunked lowendmac's error ridden RoadApple PPC myth. Only one device can talk to the processor at a time and anything on the downstream side of the bridge had a much lower bandwidth requirement than the CPU bus, so the bridge imposed no real penalty in those machines.

Bridged 030 "slow I/O bus" implementations worked well for the Mac all the way up through the PBX memory controller/bridge ASIC in the PowerBook 1400. Intel finally opened its closely held Front Side Bus equivalent to outside developers in 1997, about the same time the 1400 with its PBX bridged '030 I/O bus was discontinued.

So the whole 52/62xx Road Apple thing is pretty contentious. I'll admit that I was suckered by the LEM article myself (who would just go on the internet and post garbage? Unpossible!), the rebuttal has a couple issues itself. 68k emulation is a factor but, when running System 7.6.1 or later, it's less of a factor unless you run lots of 68k programs; anything in native PPC (and OS 8.5+ have mostly native PPC code) shouldn't suffer too much from the slightly smaller L1 cache of the original 603 vs. its enhanced successor.

To prevent further derailing this thread I'm going to do another one that goes more in-depth where I have benchmarks side-by-side of both a Performa 5320, which used the same set of bridge chips as the 52xx but ran a 120MHz 603e on a 40MHz bus, with a Performa 5420, which is PCI-based but also uses a 120MHz 603e on a 40MHz bus. Since they both run the same processor at the same speed with roughly the same amount of RAM (and the same hard drive and OS for testing), a side-by-side comparison of various benchmarks should resolve most of the debate. I will include large Ethernet file transfers, both with LC PDS and CS slot Ethernet cards on the 5320 and the CS II Ethernet card in the 5420, that should demonstrate the bottlenecks. I'm pretty confident that this test will be the most revealing because it requires data to traverse the entire logic board.