Recent content by Nammo

In the US, the N64 and Voodoo were widely available two days apart. (9/29 vs 10/1) I was there! N64 had been hyped for months, but so was Voodoo. I'd been seeing video and screenshots from both all summer long. To PC nerds in the US like me, N64 looked a lot like 20% of the power of a gaming...

TL;DR version: Development is time consuming, the creative process is often wasteful, salaries are higher than that, running a company is really expensive, especially in a high-cost city. Salaries are a small portion of the budget. I'm not sure if you're pointing to a specific game, but a...

Sorry, I didn't mean it was completely disallowed (although it was at first). I meant from the developer's perspective, there was basically no support. Nintendo just dumped their internal tools and docs on the developer, then walked away. The tools and docs were not up to devkit standards -...

We still do that, after they pass a coding test. Obviously, a high quality interview is more work than just kicking back while someone takes a premade test. Since people fail the coding test most of the time, we start with that just to save ourselves the aggravation. I've still been fooled...

The N64's RDP doesn't run programs either - its 'programmability' is by register, like the GS. Meanwhile, the N64's RCP performs programmable transform/lighting functions, like the VU. (I guess there's a distinction if you care where chip boundaries lie - but it's no different from the...

Yes, this is correct. You may be taking for granted that SSE and AVX context switching have had OS support for a while. Have you tried to use AVX on Windows XP? It doesn't work. SSE on Windows 95? Nope. MMX instructions on Windows 3.1? Same problem. Older OSes could not support these...

When I bought my first SSE-enabled CPU, I had to install drivers to experiment with the new instructions. (SSE instructions add registers, which must be virtualized by the OS, which require changes to kernel context switching before they work.) Think about the (many) virtualization features...

I'm involved in hiring and it's very frustrating. Everyone knows we're risk averse. We bypass great candidates if there are any concerns, but then still hire people who turn out awful - or worst of all, just good enough that they never get fired. And we've done our homework: We have hiring...

And think how much cheaper it would be than spending (combined with IBM) a billion of dollars and a few thousand man-years on R&D! Hmmm. They probably know something we don't. As Shifty said, if you want the same design, but multiply performance by X, you need to multiply everything else by X...

An unhappy team sounds awfully damaging. Talented people leave easily. Gripe about arrogant prima donnas all you want, but the competition will beg and bribe and promise better conditions - every week. Talented people don't want to waste their talents on a vision they disagree with. So if...

On the Jaguar, smooth and flat shaded polygons are 10-20x faster than even "fast" texture mapping like we're discussing. So, Zero 5 used textures sparingly, like all the better-performing Jaguar games. 8-bit texture mapping would be a minor gain in that context. There've been some interesting...

No, that's using DRAM as the source and the on-chip line-buffer as destination. 7 cycles is the fastest real games went, indeed using that Atari-recommended trick where src=RISC SRAM and dest=DRAM. That's 5 cycle/pixel texture-mapping blit and a 2 cycle/pixel copy blit. The RISC SRAM is...

I'm flattered you cited my post. I spent more time than I should developing hardware and software for the Jaguar. The Jaguar is a fun console for hobbyist hackers, since Atari's collapse led to everything leaking out. Not just developer docs, but internal docs, game code, boot ROM and CD BIOS...