home *** CD-ROM | disk | FTP | other *** search
- > > What do you say (Johan) about a 040 version of MiNT with XaAES?? :-0 Sounds sexy..
- >
- > Johan, are you able to recompile MiNT for 040, please?
-
- Well, I could of course of do that, but I'd be surprised if the speed increase
- was noticeable for such programs. MiNT normally doesn't do much but schedule
- the processes and handle the interface to TOS. XaAES should be mostly hindered
- by the VDI speed.
-
- > Well, I wouldn't like to test the 16+ color modes. It is really very slow :-(
-
- It can't be slower than on an original Falcon, can it?
- I mean, even if the bus access is slower it should be able to use that bus
- bandwidth for only screen data. The '040 should be a lot better at getting
- work done between the memory accesses as well.
-
- > And as I said twice before: I have just replaced the system clock 32 MHz
- > by 40 MHz.
-
- Numbers that involve the screen are always going to look very bad for an
- accelerator (on the Falcon) without new graphics hardware. :-(
-
- How about some real life benchmarks?
- Or at least ones that aren't as bad as GEMBench?
-
- For the former I'd suggest some compilation, global search and replace in an
- editor, GhostScript rendering, file unpacking/packing and a bit of (MGIF ;-)
- image processing.
-
- For the second category there's Dhrystone, Whetstone, Linpack and great bunch
- of others to be found in source code form on the internet.
-
- --
- Chalmers University | Why are these | e-mail: rand@cd.chalmers.se
- of Technology | .signatures | johan@rand.thn.htu.se
- | so hard to do | WWW/ftp: rand.thn.htu.se
- Gothenburg, Sweden | well? | (MGIFv5, QLem, BAD MOOD)
-
-