|
|---|
Asaf Ayoub United Kingdom
| | Posts 332 28 Apr 2010 15:32
| even today the OCS/ECS/AGA chipset is still a mystery.I would love to see how the chipset worked. Is it as advanced / unique as we think ? One could even write a book on the subject... A schematic of the chipset & paula would be worth gold.
| |
Marcel Verdaasdonk Netherlands
| | Posts 3976 28 Apr 2010 16:08
| Asaf Ayoub wrote:
| | A schematic of the chipset & paula would be worth gold.
|
paula is part of the chipset. ;) so what chipset are we talking about? (OCS ECS AGA?)
| |
Asaf Ayoub United Kingdom
| | Posts 332 28 Apr 2010 17:29
| i meant chipset or just paula alone would be great to see how it worked. -darn typos :-) which chipset ? any and all. mainly OCS as ECS and AGA are basicaly bolted on. still interesting to see the differences. so do you have schematic of OCS ?
| |
Thomas Richter Germany
| | (MX-Board Owner) Posts 1425 28 Apr 2010 20:44
| Joe M wrote:
|
Team Chaos Leader wrote:
| IIRC A1200 audio was the best. |
Is it a special revision of the Paula chip you're refering to here, or the mainboard maybe? I find this quite interesting. :-) I don't think anyone out there has done a serious sound comparsion between the various Amiga models and their revisions.
|
Very unlikely. The best bet is simply that the audio alias filters changed slightly.So long, Thomas
| |
Thomas Hirsch Germany
| | (MX-Board Owner) Posts 647 29 Apr 2010 01:53
| Thomas, fully correct. If you look in the schematics you'll see slight variations in the OP-Amp circuit.
| |
Team Chaos Leader USA
| | (Moderator) Posts 2094 29 Apr 2010 03:34
| I never knew why audio sounded brighter on A3000 and A1200 when compared to A500. I just assumed it was because they used different/better DACs in the newer machines. I had never mentioned it to anyone, thinking I just had a fluke machine... then one day I was speaking to Team Chaos lead Musician, Sidewinder. He just casually mentioned how his A3000 audio was so much brighter than this A500 audio. This led me to believe that my machine was not a fluke but that all A3000s & A1200s were like that. Speaking of audio flukes: According to Team Chaos Executive Vice President of the Paula Banging Dept. Damien Stewart, Different models of Amigas have the left and right switched around and different versions of documentation have the left and right switched around. So no matter how Thomas arranges the channels they will perfectly match some Amiga models and be the opposite of some others. Fun :)
| |
Joe M Norway
| | Posts 500 29 Apr 2010 05:19
| Ralph Ewers wrote:
| Interesting question. AFAIK the only known change done to Paula during its whole lifetime was the changing in housing, from DIL in the old models to PLCC in the A600, A1200 & A4000, during the move towards SMD motherboards. In that case the A600 & A4000 may have a good chance to offer better sound as well. Maybe some high frequency expert around here could evaluate whether SMD generates more or less "noise" on a PCB... (as i don't believe that the simple change in housing to PLCC instead of DIL would account for the better sound alone).Not sure if everyone heard of it: Rumors have it that the sources to the PAULA chips went lost pretty early while Commodore was still alive and kicking (some even claim that happened in the late 80's already). And the same rumors say that this was the reason why PAULA was the only Amiga chip Commodore never changed (for the housing change they could simply re-use existing binaries/matrices, but for changes like adding more channels and/or going 16bit all over they would have had to completely reinvent PAULA, using big amounts of resources and risking compatibility in the process).
|
Interesting stuff and story! I didn't know all this. Hopefully I could be able to record sound from the following Amiga models in the not so distant future: A500, A500+, A2000, (maybe A600 if a friend of mine still owns it), A1200 and A4000. Unfortunately I don't know anyone who has a working A1000 and A3000.This will be a lot of work for sure, but most likely a lot of fun too! At least it may give us an answer if Paula are "tuned" differently on the various Amiga models, or if this only happens to be a difference between the old and new generation AMIGAs. Most likely we can say that A1000/A500/A2000 belong to the old generation, but what about A500+ and A3000? They may sound different from the others, or may not. TCL claimed his A3000 sounded better than his A500, but does it sound better, worse or similar to for example the A1200? Again, this has to be confirmed by a test, or by someone who has done a "serious" comparsion. A600/A1200/A4000 would to my knowledge belong to the new generation. One question though: Could there be sonic differences between the Amigas produced by Commodore and Escom? Well, well... Maybe I go a bit too far now. I think I should stop here. ;-) Going through all the motherboard revisions as well as the different models would probably be an impossible task. Anyone out there who are willing to take the challenge? ;-) About noise: I would assume that the desktop models, where the power supply is located inside the enclosure will affect the sound in a negative direction, but obviously not if the components are well insulated. A simple way to find out if SMD generates more or less "noise" on a PCB could be to playback an empty sound sample on the actual Amiga, record the "empty" sound using a high quality RCA cable - and finally connect it to a good sound card (a high quality USB sound card for example). To me sound quality is all about how the sound is reproduced, how it actually sounds (how my ear perceive the sound, which for that matter is subjectively), not about noise frequencies. Of course, background noise is a bad thing, but a record with some background noise could also sound good. LP recordings are a good example of this. This doesn't mean that I want the Natami to have a lot of noise, but I would rather want the sound of the Natami to be as much *hi-end* as possible when it comes to sound reproduction and musicality, rather than sounding flat with no noise whatsoever. Of course I want both things (best of two worlds) if possible! I hope all these things are taken great care of when the Natami is being designed. All circuits should be separated from each other (and well isolated), so no electromagnetic interference will affect the sound in any way.
| |
Joe M Norway
| | Posts 500 29 Apr 2010 05:27
| Thomas Hirsch wrote:
| Thomas, fully correct. If you look in the schematics you'll see slight variations in the OP-Amp circuit.
|
To my understanding the Natami won't come with an original improved Paula chip (how could that be possible anyway), but with a different *sounding* one. Is the idea to *tune* this new chip to sound like for example an A1200, or is the idea to do no tuning at all? Perhaps it could be possible to add options to the early start-up control to make users choose the sound they want?For instance like this: 1. Original sound of DAC (no interpolation/tuning) 2. A500 3. A1200 4. (Eventually others) 5. ...... I know I have asked this before, but I don't think I've got an answer from anyone. Thomas: Have you decided what sound chip (DAC) you will use in the final Natami? The Samantha mainboard has a Cirrus Logic sound chip for instance. I don't believe this is a very costly chip being able to produce very good sound - since it's also to be found on rather old PC mainboards. I don't know if the chip has been improved ever since, but maybe it has. What are your plans for the Natami here? I hope the sound won't be low priority because you find it necessary to keep the final price of the Natami down to a minimum. :-)
| |
Joe M Norway
| | Posts 500 29 Apr 2010 05:56
| Team Chaos Leader wrote:
| I just assumed it was because they used different/better DACs in the newer machines. |
This is what I always believed too. Team Chaos Leader wrote:
| Speaking of audio flukes: According to Team Chaos Executive Vice President of the Paula Banging Dept. Damien Stewart, Different models of Amigas have the left and right switched around and different versions of documentation have the left and right switched around. So no matter how Thomas arranges the channels they will perfectly match some Amiga models and be the opposite of some others. Fun :) |
I have experienced that the left and right channels have been switched around on several Amiga models. Actually I think this is a rather lame mistake by the engineers (or whomever responsible), but the problem can of course easily be solved by switching the audio plugs at the back of the machine. The bad thing is, however, that it probably occured a few misunderstandings among Amiga-musicians because of this. Anyway; it is some kinda funny if you look at all the confusion it made... ;-)
| |
Steve Thomas United Kingdom
| | Posts 178 29 Apr 2010 12:29
| They say the best sound comes from a valve amplifier. But I suspect that a Natami with an on-board valve amplifier, will have a longer warm-up time than a PC's boot-up time.:-)
| |
Thomas Hirsch Germany
| | (MX-Board Owner) Posts 647 29 Apr 2010 13:20
| IDE Now after establishing a reliable two-way communication to the "Westside" CPLD the system can reach components which are located "far out" there. After the audio data which only uses one way IDE interface is now working and gives access to an attached harddisk.
 If you look close in the picture you'll see the gadgets are incorrectly drwan. This is due to the still missing line mode in the blitter. I already made some modifications to the natami-blitter enabling me to implement the line mode correctly but further changes are needed. The blitter even got its fill mode but this is not of much use without the line mode.
Frame generation .......... ECS, fixed 28MHz pixel clock SyncZorro Interface ....... preliminary version Copper .................... fully implemented Video DMA ................. fully implemented 256 color registers ....... fully implemented Sprites ................... 16bit linebuffer blitter ................... basic implementation. Block and fill mode done, line to come Video priority ............ half implemented Scandoubler ............... fully implemented Interrupts ................ fully implemented Paula DMA control ......... fully implemented Audio out ................. fully implemented VGA out ................... working DVI out ................... o PCI ....................... o (new) IDE ....................... fully implemented CIAs ...................... fully implemented Disk DMA .................. 880k and 1790k, read only Serial Port DMA ........... o Slow peripheral I/O ....... fully implemented (Joy/Mouse/Keyb/PRT/DSK/SER) PC mouse and kbd support .. o Fast RAM controller ....... o Kickstart flash logic ..... o Battery-backed up clock.... o 15k Video out ............. o 15k Video in .............. o Audio in .................. o
| |
Alexander Moon Denmark
| | Posts 33 29 Apr 2010 13:26
| WOOT! Awesome screenie! Is it via Kickstart disc or?
| |
Channel Z
| | Posts 227 29 Apr 2010 13:36
| Cool stuff! Alexander: Should be a kickstart ROM in flash memory, IIRC.
| |
Thomas Hirsch Germany
| | (MX-Board Owner) Posts 647 29 Apr 2010 13:40
| The Kickstart is in the CPU SRAM area. For now it is only downloadable via the serial debug port. Later on it will be possible soring it in the mounted flash devices when "Kickstart flash logic" (see table) is implemented.
| |
Marcel Verdaasdonk Netherlands
| | Posts 3976 29 Apr 2010 15:03
| can say one thing about it. It's Cool!
| |
Rikard Lang Sweden
| | (Moderator) Posts 86 29 Apr 2010 15:09
| Great work Thomas! Just because it's your birthday - I'll say this: "Ich liebe dich" ;)
| |
Asaf Ayoub United Kingdom
| | Posts 332 29 Apr 2010 16:48
| Impressive work ! Keep it up !Take your well deserved break.
| |
Lion 75 Denmark
| | Posts 32 29 Apr 2010 19:40
| Happy birthday, and Good work! Really looking forward to see some of the most cpu hungry scene demos fly on this thing. Another thing i want to see in action, is the gpu/3D core. Asaf Ayoub wrote:
| Take your well deserved break.
|
Eh? - What break? Please Hurry up and finish this, - i can hardly wait ;) -jk
| |
Joe M Norway
| | Posts 500 30 Apr 2010 01:54
| steve thomas wrote:
| They say the best sound comes from a valve amplifier.
|
Yes, but now you're talking about amplifiers and very hi-end stuff. ;-) I recently almost bought a STAX SRS-4040 Headphone System with Vacuum Tubes, but it was too expensive:EXTERNAL LINK Got my hands on its little brother instead, which sounds almost just as good: EXTERNAL LINK These headphones are great for critical listening and/or mixing. They are able to disclose all errors on a recording etc.
steve thomas wrote:
| But I suspect that a Natami with an on-board valve amplifier, will have a longer warm-up time than a PC's boot-up time.:-)
|
I assume not many people would want a solution like that. ;-) I do not expect the Natami to use specialized hi-end audio components, but I expect it to have something better than this:EXTERNAL LINK Since it's already been said that the Natami will have 24-bit output, it can't be that bad. I'm just curious in what league it will be. 24-bit is only a number, and doesn't necessarily mean good sound. If the Natami could have a sound chip similar to Soundblaster Audigy 2 ZS in quality for example (which is now eight years old), I would be happy. You can get these cards for about 30 USD these days, so hopefully it wouldn't be out of reach to adapt a similar chip to the Natami as well. I guess only Thomas can tell... EXTERNAL LINK |
|