home *** CD-ROM | disk | FTP | other *** search
- Path: news.vol.it!news
- From: bizzetti@mbox.vol.it (Fabio Bizzetti)
- Newsgroups: comp.sys.amiga.programmer
- Subject: Re: CHIP RAM speed test resul
- Date: 26 Mar 1996 01:34:29 GMT
- Organization: Video On Line
- Distribution: world
- Message-ID: <5827.6659T112T770@mbox.vol.it>
- References: <4j6jv0$1im@serpens.rhein.de>
- NNTP-Posting-Host: molcl14.vol.it
- X-Newsreader: THOR 2.22 (Amiga;TCP/IP)
-
-
- >bizzetti@mbox.vol.it (Fabio Bizzetti) writes:
-
- >>It's not "all or nothing". Using the same chips speed, but having dual
- >>access would already mean to avoid the syncronization with 7Mhz chipram bus
- >>clock. And, whatever is the amount of DMA activity, the CPU wouldn't be
- >>slowed down.
-
- >Obviously this is not true. Since RAM of a given speed just allows
- >that many accesses per second you will have to slow down accesses.
-
- As far as I recall my A1200 chipram is 80ns; while it's exploited as if it was
- only 140ns. The rest of ns is wasted because Alice holds control of everything,
- and Alice is slower than the CPU.
- Since you didn't get what I meant last time, I'll try to explain my thought
- better: if you put 10ns ram in your 68030 board, it doesn't mean that it will
- be exploited as 10ns. It will be exploited as your ram controller allows, 60ns
- or 70ns commonly. Well, the chips inside my A1200 ("chipram") are much faster
- than they're supposed to be, but Alice brakes the CPU to gain from this speed
- because in fact Alice is chipram's controller.
-
- >It is clear that you can do better than current 7MHz chip RAM but
- >that's because RAM is faster nowadays.
-
- It would be sufficient to modify the interface CPU <->chipram<-> Alice.
-
- >Yes. That's why "dual ported RAM" is nonsense.
-
- Come on.. I didn't mean to simply change the chips!!!!
- There must be a circuit to exploit the new chips' nature, and getting a fast
- SIMM and multiplexing data is perfectly fine, allowing practical dual access,
- but (perhaps) could be more complicated at the end than real dual access ram
- (ofcourse with the circuit to use it).
- Maybe the "ram controller" that you wish (to multiplex data) would be more
- expensive than 2Mb of real dual access ram, this is to be examinated.
- The implementation is is not our problem though: we know it can be done, then
- if a solution is cheaper than another, it doesn't change the final result.
-
- Everything is good to improve performances, and AGA has already a very good
- 22Mb/sec bandwidth to RGB, giving another 22mb/sec CPU-chipram would be *nice*.
-
- A bit useless if the CPU is slow of course, but a 40Mhz/50Mhz 68030 (finally
- standard) is well worth of these 22mb/sec.
-
- > We already have "dual port RAM", just the multiplexing is done outside the
- > chip.
-
- This is the problem: it's performed by the 7Mhz (bus clock) of Alice.
- AGA is really nothing newer than OCS, the differences are more apparent
- (palette, 8bitplanes) than physical (just 32bit interface implementation
- and fastpage modes).
-
- AGA can be improved a lot though, the AgaEXTENDER would make it (visually)
- better than any SVGA out there (at least for games and Video/MultiMedia
- applications), and having (still not needing to modify the chipset) a dual
- access ram (any kind, any solution: it has only to "work" ;) ), syncronous
- (to semplify the circuit) 24Mb/sec (22Mb/sec is due to vertical blank, but
- the limit is 24Mb/sec: well, the AgaEXTEDER doesn't lose these 2Mb, that are
- available for audio) would make the AGA *great*.
-
- IMO, instead of 2+4Mb (Walker) it would be much better 2+2Mb, but with an
- access speed in chipram of 24Mb/sec. Ofcourse, 2+4Mb would be still better O:)
-
- Just think, 24Mb/sec is nearly the speed of FastRam: you would avoid the
- fastram->chipram copy and draw directly in chipram, you would avoid c2p
- (thanks to the AgaEXTENDER), it would allow to use 2 byteplanes (or 1 word)
- for texturemapping+shading effects, the AgaEXTENDER would allow a lot of great
- effects for free (zoom, flip, 4 playfields, hardware handled parallax, hardware
- scroll, YUV conversion, antialiasing, any mode/resolution programmable,HiRes
- copper, great audio).
-
- AT could sell this solution as single, to SetTopBox / MultiMedia companies,
- thus financing the Amiga indirectly.
-
- >>It seems that Escom/AT considers the actual AGA chipset as perfect, or
- >>anyway "adequate". That's why the Amiga games market doesn't exist anymore.
-
- >Modern games do not need smart chipsets.
-
- Please believe who (probably) made more games than you: what you said is wrong.
-
- > They need geometry engines,
-
- A skilled assembly programmer can avoid multiplications/divisions and all
- these things that normal programmers learn from standard books.
-
- If AT wanna give me voice, allowing the AgaEXTENDER for A1200 and also new
- chipmem interface for the Walker, I'll give publicly all the 3D tricks you'll
- see in my next demo (for StarFighter and VirtualRally), allowing all the
- skilled Amiga coders to exploit the AgaEXTENDER and the 68030 to the limit.
-
- It would mean to blow the PC away.
-
- >texture mapping and a fast chunky framebuffer.
-
- Yes, but dont forget that it's not so important for games, when i.e. AB3D-II
- (not talking about BreathLess..) are so slow and bad thought (also if
- optimized), where having the best SVGA card would make them faster of 5%-15%
- max.
-
- >I.e. a Pentium and PCI graphics board.
-
- They cost a lot, the PCI GfxBoard isn't flexible (it gains all from Pentium's
- MIPS, that are quite more than Walkers' ones), the 3D cards *still dont exist*,
- they will cost a lot and they aren't flexible (read: only textmap polygons,
- a bit wasting of money for a general purpose and versatile *computer*).
-
- >--
- > Michael van Elst
-
- >Internet: mlelstv@serpens.rhein.de
- > "A potential Snark may lurk in every tree."
-
- Regards.
-
- /-----------------------------------------------------------------------\
- | Fabio "Maverick" Bizzetti - bizzetti@mbox.vol.it - Maverick* at IRC |
- | The maker of "CyberMan" and "Virtual Karting" |
- | working on "VirtualRally" & "StarFighter" |
- \-----------------------------------------------------------------------/
-
- PS: will I win the "shortest signature of the year" prize this time? ;)
-
-