Search results

Sorry about the PSU fan, but good you found out before it was too late! Thanks for doing that, I added the numbers into the spreadsheet. It is actually *very* interesting to me that you can get such high fps with the video-chip as-is. I had a quick look in the developer notes of the...

Disabling the card, wouldn't that be exactly like running without it?

Very nice! Good to see that the civic video chip can push up to 39.8 fps, which is likely the bottleneck with the G3 in it. I hope we can get a 6100/66 to compare with and see how much that accelerator card is helping out on that same video chip.

That's interesting! OK I will need to fire up my Pentium 100 and do some comparisons eventually. I have only used it to watch some demos, and my memory of my friends 486/Pentiums might have been too colored by how the games performed, although I recall thinking Windows felt quite snappy...

Cool! Thanks for doing that :) I thought it would do better than linear as it is the 603e so should be able to pair instructions better.

It should be amazingly powerful compared to a Performa 400. The question is, how was it measuring up against its competition, which by 1995 was Pentium/top 486. L2 cache miss is roughly 32 cpu cycles for RAM. And valkyrie write buffer is seemingly available again after 8 cpu cycles. It is...

This is true, but exactly what is causing the quite substantial slow down I'm not 100% sure yet. Here is what I do know: the 603 LSU stalls on cache miss, meaning no more loads or stores can happen until the cache miss has been fully resolved. The 603 can still do FPU and ILU instructions, but...

Thank you for doing it, looking forward to see the results! :)

It seems someone deleted the instructions in the spreadsheet, I added them back. Posting here as well: Quake - Get the 2nd download: https://download.macintoshgarden.org/games/Quake109.sea_.bin - Install 1.08, then update to 1.09. - Settings: set to 320x200, run with/without 2x scaling...

This attached paper is also interesting. It seems they didn't consider *expensive* cache misses at all. Similar to what was speculated earlier in the thread. Was the 603 badly designed, or was it just not used in the way it was designed for? One could probably argue both ways. I don't know how...

Their methodology was essentially: measure the real test, take note of ~2.5% cache misses, then trace test without considering the cpu state and assume 2.5% of LSU is cache miss, and that the cost of the cache miss is the cost of looking it up in an L2=1MB. Had they had as a requirement that...

I have setup a spreadsheet to benchmark Mac computers on Quake: https://docs.google.com/spreadsheets/d/1ndaq6qB-cMrCBbtbENK5MjTWr4g-nnw6yWjLqYM2ARM/edit?usp=sharing I would be very interested in seeing what FPS you would get on your mac.

I found an interesting article for the 603: "PowerPC 603, A microprocessor for Portable Computers" - written by the PowerPC 603 design team. - "As this article shows, its design focuses on meeting these challenges and achieving an optimum balance between power and performance for the portable...

It is not perfect, I give you that. And Doom might be the better benchmark for these machines, or at least to compliment. I think PowerPC can beat Pentiums on Quake 1. A 604 should be able to win over Pentium (at same MHz) on Quake 1, as its fpu can complete the div in parallel, while it still...

This is an interesting source. Thanks! The 603e should not be confused with the 603. While they are very similar, they are different enough that they should be treated separately: - 603e: stores are faster (2x throughput) - 603e: the SRU can do some integer work, so is "~2 ALU" (like the 68060...

Alright let's talk! :D Quake 2 and esp 3 are irrelevant for the 603. They were not made for that type of CPU, came much later and were meant for more powerful CPU:s. I am not suggesting to use Quake 1 to measure "any CPU", I'm thinking specifically for the 603 and its "competitors". You are...

Hey @Snial! I think it is also the bus speed that impacts quite a bit here given the many load and stores, the 601 that is handicapped by not having dual issue gets 7.6 fps (although this number is not directly comparable as the timedemo was on a different map but it is in same ballpark)...

I now believe Quake FPS is the best way to answer this question. Benchmarks - we saw it, they are not reliable (not representative work, depends as much on the compiler as the cpu...). I've run Quake now on my 5200 (note: one must first start a game, then open console and write timedemo demo2...

Yes! So this thread here is really me wanting to fix my performa to run with the CRT, but I don't know how to power the analog board with an ATX - if at all possible. And frankly, I'm a bit worried about repairing a PSU (electrical shock, fire, etc). Would love to get the CRT back... and a way...

Just checked the manual, and uh hehe.. I recalled incorrectly... only instructions that can essentially skip the first stage in the FPU are equally fast in single precision, otherwise they do take one extra cycle. (And way worse for already expensive instructions like fdiv vs fdivs). Pg36...