English Amiga Board


Go Back   English Amiga Board > Search Forums

Showing results 1 to 40 of 104
Search took 0.01 seconds.
Search: Posts Made By: chb
Forum: Coders. Asm / Hardware Yesterday, 20:17
Replies: 830
Views: 28,146
Posted By chb
Maybe also due to prefetch: 8086 can fetch more...

Maybe also due to prefetch: 8086 can fetch more instructions and more freely, so it can fill up its prefetch queue with several instructions while an instruction with a higher cycle count is...
Forum: Coders. Asm / Hardware 17 November 2018, 14:31
Replies: 830
Views: 28,146
Posted By chb
Yeah I know, I just had in mind everything that...

Yeah I know, I just had in mind everything that you could stick into the PCMCIA slot, my comment was only related to general timing, not specifically to refresh.

Roondar: Very interesting result!...
Forum: Coders. Asm / Hardware 17 November 2018, 13:59
Replies: 830
Views: 28,146
Posted By chb
Wouldn't fast ram also need refresh cycles? Maybe...

Wouldn't fast ram also need refresh cycles? Maybe UAE simply does not emulate those, as probably there isn't any software relying on a specific fast ram timing (which could also vary a lot between...
Forum: Coders. General 16 November 2018, 17:23
Replies: 13
Views: 433
Posted By chb
I'd even bet that if you'd substitute 0x8000 in...

I'd even bet that if you'd substitute 0x8000 in BPLADAT by 0xC000, 2x1 would come for free. But it's very likely still slow as hell. :D
Forum: Coders. General 15 November 2018, 15:56
Replies: 13
Views: 433
Posted By chb
That will very likely be awfully slow compared to...

That will very likely be awfully slow compared to the classic blitter c2p. But no harm in trying, though you may want to use BLTBDAT for the pattern instead, even if BLTADAT might give some funky...
Forum: Coders. General 12 November 2018, 15:29
Replies: 7
Views: 210
Posted By chb
You can use blitter merge (two blits with shift...

You can use blitter merge (two blits with shift by 8 left resp. right for one source, mask 0xFF00 and 0x00FF) to get the following intermediate matrix:

a-b-i-l
e-f-o-p
c-d-m-n
g-h-q-r

then...
Forum: Coders. Asm / Hardware 12 November 2018, 10:54
Replies: 830
Views: 28,146
Posted By chb
I think this is a misunderstanding; the...

I think this is a misunderstanding; the Archimedes has 4x the Amiga 500's memory bandwidth, but its DRAM chips are not much faster rated (I think 120 ns versus 150 ns). The advantage comes from two...
Forum: Coders. Asm / Hardware 08 November 2018, 01:41
Replies: 1
Views: 246
Posted By chb
There is no speed difference between EHB and...

There is no speed difference between EHB and ordinary 64-color-mode when both are displayed on AGA. Both can be faster than EHB on OCS/ECS, tho, when the faster transfer modes (fetch modes) of AGA...
Forum: Coders. Asm / Hardware 31 October 2018, 19:01
Replies: 830
Views: 28,146
Posted By chb
True, but in my eyes it's not a design very well...

True, but in my eyes it's not a design very well suited for scrolling if your system with 4x the bandwidth of the Amiga after display dma and scrolling is left with about the same bandwidth... not to...
Forum: Coders. Asm / Hardware 31 October 2018, 18:35
Replies: 830
Views: 28,146
Posted By chb
AFAIK it's slightly less (25.6MB/s), as it's...

AFAIK it's slightly less (25.6MB/s), as it's using page mode with a 2-1-1-1 pattern, but still a lot.
Forum: Coders. Asm / Hardware 31 October 2018, 18:32
Replies: 830
Views: 28,146
Posted By chb
Correct for raw transfer bandwidth, but for...

Correct for raw transfer bandwidth, but for scrolling you cannot write 32 bit words to non-aligned addresses, so you need a) load with ldm, b) save the bits that will be shifted out of the longword...
Forum: Coders. Asm / Hardware 31 October 2018, 17:21
Replies: 830
Views: 28,146
Posted By chb
and d) your code to move/bitwise scroll the data...

and d) your code to move/bitwise scroll the data has also to come from memory, if there's no cache present. It's simply a waste of memory bandwidth for something that can be done with little effort...
Forum: Coders. Asm / Hardware 27 October 2018, 22:40
Replies: 830
Views: 28,146
Posted By chb
Well, my point was that in terms of absolute...

Well, my point was that in terms of absolute performance (not per clock) ARM CPUs were never among the fastest of their time. I guess that wasn't their goal anyway.

All true - and the ARM2 is...
Forum: Coders. Asm / Hardware 27 October 2018, 18:42
Replies: 830
Views: 28,146
Posted By chb
If you compare ARM processors to other...

If you compare ARM processors to other manufacturer's in CPUs from the same time frame, they are not particularly fast. The 1987 ARM2 is not faster than the 1987 68030/80386, which came at higher...
Forum: Coders. Asm / Hardware 27 August 2018, 12:15
Replies: 830
Views: 28,146
Posted By chb
Well, not really. You could a) fit several...

Well, not really. You could a) fit several sequential accesses in one slot if your mem is fast enough (like AGA's Lisa does, two 32-bit accesses) and b) it is not a strictly interleaved design, the...
Forum: Coders. Asm / Hardware 19 August 2018, 23:22
Replies: 830
Views: 28,146
Posted By chb
double post, please ignore

double post, please ignore
Forum: Coders. Asm / Hardware 19 August 2018, 23:11
Replies: 830
Views: 28,146
Posted By chb
I see the problem to get reliable benchmarks for...

I see the problem to get reliable benchmarks for such old hardware, but The ones given in the BYTE article are just nonsense for your comparision, they measure BASIC and disk/file access between...
Forum: Coders. Contest 16 August 2018, 21:33
Replies: 328
Views: 17,896
Posted By chb
If you have a lot of memory to waste, you could...

If you have a lot of memory to waste, you could also go full brute force and store 16 shifted versions of your bg image, Atari-style. :) Then all you'd need to do is setting the bitplane pointers....
Forum: Coders. General 16 August 2018, 15:25
Replies: 120
Views: 8,127
Posted By chb
No idea if it was the fastest (how would you...

No idea if it was the fastest (how would you measure that, btw?), but I do not know any game that was faster at that level of detail. It's an ST port for sure, but a) Thalion showed their competence...
Forum: Coders. General 14 August 2018, 16:24
Replies: 120
Views: 8,127
Posted By chb
Yes, sure, did I write something else? The cpu...

Yes, sure, did I write something else? The cpu can access the memory at max every 2nd memory cycle (once in 4 cpu clock cycles), 16 bit wide. But movem from register to memory is minimal 8 + 4n...
Forum: Coders. General 14 August 2018, 15:06
Replies: 120
Views: 8,127
Posted By chb
@roondar: Nice find with Alpha Waves, thanks. ...

@roondar: Nice find with Alpha Waves, thanks.

I'd also say that C just does not give you enough control for optimizing the time critical inner loops. If portability is not important for you, I'd...
Forum: Coders. General 11 August 2018, 18:32
Replies: 120
Views: 8,127
Posted By chb
Ah, interesting. As said, the fastest (and most...

Ah, interesting. As said, the fastest (and most complex and inflexible) way probably would involve writing individual code for every object shape, if you do not plan to dynamically modify them. Which...
Forum: Coders. General 10 August 2018, 13:27
Replies: 120
Views: 8,127
Posted By chb
Yep, that's what I wanted to say. As symmetries...

Yep, that's what I wanted to say. As symmetries around the origin result in same abs values for the vector elements, they are therefore beneficial for this trick. If you have other models with same...
Forum: Coders. General 10 August 2018, 11:33
Replies: 120
Views: 8,127
Posted By chb
Hmm, either I fail to make myself clear or my...

Hmm, either I fail to make myself clear or my idea is utter nonsense, both of which is totally possible :). Maybe to be clear: The idea is avoiding multiplications. If your multiplications are fast,...
Forum: Coders. General 10 August 2018, 00:57
Replies: 120
Views: 8,127
Posted By chb
My idea was mostly about reducing the number of...

My idea was mostly about reducing the number of values the coordinates can take. I was assuming you are working always with the original coordinates, or do you rotate the rotated coordinates again,...
Forum: Coders. General 09 August 2018, 23:28
Replies: 120
Views: 8,127
Posted By chb
No, that one relies on some properties of the...

No, that one relies on some properties of the binary representation of floating point numbers and makes only sense when there's a reasonably fast fpu present. On an 68000 it would be much slower...
Forum: Coders. Contest 06 August 2018, 16:40
Replies: 328
Views: 17,896
Posted By chb
If you're using 32 color bobs, it gets really...

If you're using 32 color bobs, it gets really quite tricky, and you'd need to start the blitter twice as often (unless you are not using interleaved bitplanes, but I guess you do anyhow). So quite...
Forum: Coders. Contest 06 August 2018, 14:38
Replies: 328
Views: 17,896
Posted By chb
Just a little idea: As I understand it, you're...

Just a little idea: As I understand it, you're using a 5 + 3 bitplane setup, with a palette chosen in a way that emulates dual play field mode. Then you're scrolling the 3 background planes with the...
Forum: Coders. Asm / Hardware 05 June 2018, 15:24
Replies: 31
Views: 1,920
Posted By chb
Yep, it's probably useful only for a quite...

Yep, it's probably useful only for a quite restricted number of scenarios: no competing dma (borders), cpu busy with other tasks like muls and shifts that do not require a lot of memory cycles, or if...
Forum: Coders. Asm / Hardware 05 June 2018, 00:16
Replies: 31
Views: 1,920
Posted By chb
If you optimize for 68k, you could also try to...

If you optimize for 68k, you could also try to use the blitter. I was theorizing about it some time ago here: http://eab.abime.net/showpost.php?p=1172367&postcount=248 (didn't know it has a fancy...
Forum: Coders. General 31 May 2018, 23:57
Replies: 7
Views: 715
Posted By chb
I'm not completely sure what you want to gain by...

I'm not completely sure what you want to gain by this technique. For a color change every 8 pixels you can not use more than 16 colors, so max 4 bitplanes. Also you need to change color 4 for every...
Forum: Coders. Asm / Hardware 03 May 2018, 09:29
Replies: 14
Views: 1,581
Posted By chb
There's an undocumented feature that might be...

There's an undocumented feature that might be handy:
http://eab.abime.net/showthread.php?p=1145422

Basically it gives you a 5-bitplane mode where a 1 in the 5th plane gives you always color 16,...
Forum: Coders. General 24 April 2018, 14:07
Replies: 8
Views: 2,356
Posted By chb
AFAIK CAOS was written (according to Amiga Inc.'s...

AFAIK CAOS was written (according to Amiga Inc.'s spec) by a third-party developer anyway, who did not kept schedule/wanted to renegotiate contracts after Commodore bought Amiga (there are different...
Forum: Coders. Asm / Hardware 14 April 2018, 21:36
Replies: 7
Views: 708
Posted By chb
@robinsonb5: That's a nice method, thanks for...

@robinsonb5: That's a nice method, thanks for sharing!



Hmm, I'd think a HAM8 fade on AGA would be same speed or faster than a HAM6-fade on OCS (both lowres). HAM6 on OCS: ~40,000 chipmem slots...
Forum: Coders. General 07 April 2018, 18:18
Replies: 587
Views: 48,332
Posted By chb
Again: How do you handle this in a multitasking...

Again: How do you handle this in a multitasking environment? Do you expect application programmers to tailor their code (that's maybe written in C or Pascal) to this 32k ram?



That's clear -...
Forum: Coders. General 07 April 2018, 15:48
Replies: 587
Views: 48,332
Posted By chb
How do you want to do that? How do you ensure...

How do you want to do that? How do you ensure that applications use that 32k for their most time critical routines? How do you handle multitasking? Swapping between those 32k and chipram at task...
Forum: Coders. General 07 April 2018, 15:27
Replies: 587
Views: 48,332
Posted By chb
Well, if you have the DRAM technology of that...

Well, if you have the DRAM technology of that time and a 16 bit wide bus, you do not get higher memory bandwidth without considerable effort, like multiple memory subsystems... And 32k of fastram...
Forum: Coders. Asm / Hardware 15 March 2018, 09:43
Replies: 21
Views: 2,175
Posted By chb
Yep, it's not related to blitter c2p, in fact,...

Yep, it's not related to blitter c2p, in fact, britelite's demo wolf3d was using blitter c2p form the beginning, but the rotated chunky buffer came later. AFAIK, he rotates it back to normal layout...
Forum: Coders. General 09 March 2018, 14:46
Replies: 60
Views: 6,736
Posted By chb
Sounds like some seriously crazy hack! :) It...

Sounds like some seriously crazy hack! :) It doesn't involve some fiddling with the blitter modulos by any chance?



I guess we have to just wait for any answer... :D
Forum: Coders. General 07 March 2018, 23:29
Replies: 60
Views: 6,736
Posted By chb
Stephane is still active on the Sega16 forum, so...

Stephane is still active on the Sega16 forum, so should be easy to contact him there:

http://www.sega-16.com/forum/member.php?141357-gasega68k

I'd guess blitter-assisted c2p should be fast...
Showing results 1 to 40 of 104

 
Forum Jump

All times are GMT +2. The time now is 01:21.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2018, vBulletin Solutions Inc.
Page generated in 0.08088 seconds with 10 queries