Socket 4 with VLB

[fsmith2003]

Curious. Are there any socket 4 motherboards with VLB slots? I don’t think I have ever came across any.

 

[Jackson]

A friend I know who used to work at DEC received a Socket 4 Pentium board shoved in a rotten XT clone case, and it had VLB slots. They do exist.

 

[lowen]

There were a few, and none worked very well. Back in the VCF archives you can find http://www.vcfed.org/forum/showthread.php?6587-The-Worst-Pentium-Motherboard-Ever-Made

After all, the VLB is essentially the 486 bus brought out to a slot, and the Pentium bus is rather different from the 486 bus.

Another discussion thread on a different forum, talking about a Socket 5 with VLB: https://www.vogons.org/viewtopic.php?t=30689

I have personally laid eyes and hands on one of these, years ago, and it was an absolute dog; the 486DX-100 ran much smoother with VLB cards than the Pentium did. All of the VLB-bearing Pentium boards I or any of my friends had have long since gone to the recycler. I never owned one; by the time I was ready to upgrade my 486DX50 (not DX2, straight DX50) the Pentium-with-VLB as well as Socket 4 in general was already old, and I upgraded to a Socket 5 with PCI slots running an AMD K5-90.

The most difficult card to get running in my DX50 was a VLB SCSI card; I had SCSI drives instead of IDE in my DX50. I think I had the adaptec AHA 2842A; in looking at a few eBay listings, that looks familiar. It was a definite upgrade from the ISA ESDI setup I had with a Maxtor 690MB full-height drive.

EDIT: On second thought, it might have been an Ultrastor 34F that I had......

Further EDIT: the Vogons Wiki page on Socket 4 motherboards: http://www.vogonswiki.com/index.php/Socket_4_Motherboards

 

[Jackson]

A friend I know who used to work at DEC received a Socket 4 Pentium board shoved in a rotten XT clone case, and it had VLB slots. They do exist.

Turns out it was just a regular 486 motherboard that housed a DX4/100. Boo hoo!

 

[Unknown_K]

Early socket 4 boards were pricey as hell and you either got EISA/ISA or PCI on later OEM models. I heard of socket 5 VLB mentioned but never seen one, would have been too expensive for home builders and PCI was a better match anyway.

 

[Anonymous Coward]

Early socket 4 boards were pricey as hell and you either got EISA/ISA or PCI on later OEM models. I heard of socket 5 VLB mentioned but never seen one, would have been too expensive for home builders and PCI was a better match anyway.

Believe it or not, there was a time when the Pentium was available when people were still highly skeptical of PCI. Most of the Pentiums available during this time were either MCA, EISA, VLB or even plain old ISA! ALR and Acer were two of the big OEMs to release Pentium systems with VLB slots, and I think even IBM did through their Ambra line (using the Acer board).

VLB actually works just fine with a Pentium. The board from my old post just had some bad solder joints, but I have a new VLB Pentium board and it will take any VLB card I throw at it, and performs quite well. Even AMI made a board with a VLB slot!

The performance issues with VLB on Pentium arises mostly on those dreadful VIP combo boards.

In my opinion the real failure of VLB came down to the fact that it was simply too long (wasted too much PCB), and its features were too limited compared to PCI. The basically unused VLB 2.0 spec would have solved many of the problems, but it came out too late to matter.

 

[Unknown_K]

The problem with VLB was the speed of the BUS was tied to the FSB of the CPU (and writing to the cards meant you couldn't read from CPU cache at the same time as they shared the same bus). It was hard making cards work at greater then 40mhz speeds (people had problem with the 486DX50's) so once Pentiums came out at 60 and 66mhz and greater VLB was in trouble. PCI's claim to fame was a separate clock from the CPU and you could use more then 2-3 cards.

I don't think wasted PCB space was too much of an issue in the early days since most PCI cards (other then Ethernet which was rare in VLB and sound cards that were nonexistent in VLB) in the beginning were pretty large. VLB cards seemed shorter to make up for a longer length.

Are we now talking Pentium 1's in general or Socket 4 which was the original question? I have P1's in EISA and MCA.

 

[GiGaBiTe]

PCI's claim to fame was a separate clock from the CPU and you could use more then 2-3 cards.

There were quite a few 486 boards that ran PCI off the CPU FSB clock. There were also a few very cheap Pentium/AMD k5/6 that did this as well.

For the majority of the PCI bus lifetime, it didn't have an independent clock source and was run off a clock divider from the FSB, memory or AGP clock, meaning that the slot could be under or overclocked depending on the CPU used.

Cyrix and some AMD CPUs were notorious for causing PCI issues due to their weird bus speeds. Common PCI bus divider options were 2/3 or 1/2, so a Cyrix with an 83 MHz bus would run the PCI slots at an alarming 41 or 55 MHz.

 

[lowen]

Believe it or not, there was a time when the Pentium was available when people were still highly skeptical of PCI. Most of the Pentiums available during this time were either MCA, EISA, VLB or even plain old ISA! ...

As I recall, the biggest advantage VLB had over PCI back in the day was that you didn't lose an ISA slot for each VLB slot, but you did loose an ISA slot for each PCI slot (except the rare boards where one PCI and one ISA could coexist). It took a while for common cards to become available in PCI versions, most notably sound cards, where one of the first PCI sound cards was the Ensoniq AudioPCI (later bought out by Creative and remarketed in the Sound Blaster line). I think I still have my original pre-Creative acquisition AudioPCI.

These were the days before integrated I/O chips and such, so you needed a slot for your VGA, and you needed a slot for your Ethernet, and you needed a slot for your Sound Blaster, and you needed a slot for IDE/Floppy, and another for parallel and serial (later combo I/O cards freed up one slot). Common PCI/ISA boards in the early days left you with a cramped feeling on the ISA side.

The advent of ATX and most of the I/O being on the motherboard changed all that, and by then most I/O was available in PCI form.

 

[VintageUniverse]

I believe some Socket 4 based Packard Bell machines featured VLB on the rise card, correct me if I am wrong. I might be referring to the late 486 DX2 Packard Bells.

 

[Anonymous Coward]

The problem with VLB was the speed of the BUS was tied to the FSB of the CPU (and writing to the cards meant you couldn't read from CPU cache at the same time as they shared the same bus). It was hard making cards work at greater then 40mhz speeds (people had problem with the 486DX50's) so once Pentiums came out at 60 and 66mhz and greater VLB was in trouble. PCI's claim to fame was a separate clock from the CPU and you could use more then 2-3 cards.

I don't think wasted PCB space was too much of an issue in the early days since most PCI cards (other then Ethernet which was rare in VLB and sound cards that were nonexistent in VLB) in the beginning were pretty large. VLB cards seemed shorter to make up for a longer length.

Are we now talking Pentium 1's in general or Socket 4 which was the original question? I have P1's in EISA and MCA.

It's true that VLB had issues when cards operated at 40 and 50MHz, but the same was also true for PCI. IF you look at people who overclocked Socket7 systems to 75 and 83MHz for example, a lot of their expansion cards got flakey.
I can't speak for all of the chipset manufacturers, but OPTi was able to solve this problem by making VLB run at half the CPU clock. You can confirm this in their sales literature. I highly doubt any of the Socket4/5 VLB Pentium boards ran their VLB at the full CPU clock. So on a Pentium, VLB is not really a local bus (pretty amusing, eh?).
I don't believe the number of PCI slots was an issue at the consumer level at the time. Almost all early PCI boards had 3 slots like VLB, and if there was a fourth it was gimped. In order to get additional slots I believe a PCI to PCI bridge was required. I can only assume that similar glue logic could have also been implemented on VLB if desired. Back then people normally used at most 1 slot for VGA. Maybe 2 if they had ethernet (pretty uncommon until broadband).

Also, if you look at many of the VLB graphics cards from 1994 or so, you can see there are large sections of blank PCB. I suppose they could have made the cards really low profile, but then you're still wasting valuable real estate on the motherboard. Those 1/2 length AT boards just aren't possible when you have even 1 VLB slot present.

 

[mR_Slug]

I've done a bit of research into socket 4 Pentium and PCI. For all the problems you may have with VLB, PCI in the early
days (1993) was a mess. It wasn't really until 1994 that most of it's problems were ironed out.

As far as I can tell, in the early days Intel was the only company providing PCI chipsets for the 486 and Pentium. The
Intel Saturn and Aries 486 chipsets were not particularly fast. If you wanted PCI and a Pentium you had one option, the Intel 82430LX (Mercury) chipset. I think the PCI based Intel "xPress Extended" was a later product.

The early Mercury chipset has a problem with the write-back cache of Pentium. This slows down PCI/ISA systems like the early Batman boards. It is particularly bad on the PCI/EISA systems. By Oct 04, no vendors were showing off PCI/EISA systems (InfoWorld Oct 4 1993 p8)

Intel seem to be fixing this by late October (Computerworld Oct 25, 1993 p45) Major manufactures have delayed shipping PCI systems due to this problem. Intel appears to have fixed this issue by Nov 29, when it release a new chipset revision (InfoWorld Nov 29, 1993 p1). Bus mastering, related to this issue is now fixed.

Other problems:
End of November, Intel redefines the Pentium spec to use 5.6v for the Pentium 66Mhz. A Non-trivial change in M/B design, further delaying Pentium adoption (InfoWorld Nov 29, 1993 p10)

Windows NT (3.1) Lacks PCI drivers as it was not available when MS shipped NT, "We didn't put it in because there wasn't any hardware to test with it" - Stork, "We just haven't been able to get manufactures to ship us PCI based systems so that we could test them." - Stork. MS has driver, but its not available yet. (InfoWorld Nov 22, 1993 p2).

There is also a problem with PCI-to-PCI bridges. From the later 450KX/GX datasheet: "Finally, if a P2P device in use is not fully compliant with the PCI 2.1 specification, the system is exposed to unresolvable conflicts between multiple bus masters issuing transactions attempting to cross between the hierarchical PCI busses. To eliminate the possibility of a resulting livelock failure, the system must operate with CPU-PCI write posting disabled. This will degrade the performance of outbound traffic such as graphics, but will not adversely affect the performance of bus mastering I/O devices." AFAIK this is describing shortcomings in PCI 2.0.

Conclusion:
PCI was very dependent on the 430LX chipset, and the early revisions have some serious issues partially caused by driver incompatibility, but also inherent in the chipset. The PCI/EISA variant particularly suffered. In addition, because of the easy adoption of VESA on the 486, manufactures were more inclined to build VESA addon cards. A PCI/ISA system built on the Intel Batman platform, side stepping the EISA issue, would have suffered from a lack of high end EISA add in cards. Everything fast would have to be on the PCI bus. However this has problems with bus mastering and the very useful PCI-to-PCI bridge, used on many later cards wont work. So it would be difficult to recommend PCI in 1993. It just wasn't an established technology. EISA would be the standard in the server world for at least a year.

An analysis of the the relevant datasheets.
If this seems confusing it's because it is. Intel keeps redefining the the spec. For example the 82378IB southbridge originally supports 4 PCI masters, then it's only two, but the 82378ZB supports 4, but then it's 6. See the notes section for a summary.

TL;DR
In short buy a VLB based 486 in 1993, and wait till '94 for a Pentium (as most did). Alternatively buy a Compaq or ALR in 1993.