Q:Commodore A3640 tinkering - Watch out I have boing balls!

  • Thread starter Thread starter arnljot
  • Start date Start date
  • Replies Replies 83
  • Views Views 13073
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Before I look @ your PM's, yeah the Cr board has a battery. It's the Coin type like the A4KT & it only has DIMM slots for Fast Ram. The Chip Ram is surface mounted on the Mobo & the 030 is also on the Mobo! There's also a 33Mhz PLCC FPU fitted but it runs @ 25Mhz as per the CPU X-Tal. There is a mounting for another X-Tal to run the FPU at a faster speed, though the position is naked atm! :mrgreen:

Charlie
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

I look forward to hearing from you! Waiting with xitement! :P :woot:
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

TheCorfiot said:
Kin Hell said:
The Mobo always takes it's speed from the Crystal on the Mobo.
The INT/EXT jumpers merely tell the Amiga which Speed crystal to use for the CPU/FPU, NOT the Chipset! :wink:
Kin
Click to expand...

I absolutely agree with Kin.

The Mobo conatains 2 Oscillators
1 - 50MHz which is used in "INT" mode to drive the 3620/30 CPU card which has no oscillator on board or "EXT" which means the relevent CPU Slot pins will hold the clock generated by the CPU card's on board oscilator, for ex, the A3640.
2 - 28.37516MHz (for PAL) this Oscillator is used for the chipset timing & is independant of the CPU oscillator.
TC :mrgreen:
Click to expand...

Hmmm.. perhaps I'm confused because I was thinking of the A3000 (http://groups.google.com/group/comp.sys ... d72b41260c), but after re-reading the master's (Dave Haynie) explanation, I don't see why the custom chips on the A4000 would be clocked differently than those in the A3000. Perhaps they're "pinned" at 25Mhz, but the overclocked A3640 benchmarks show RAM bandwidth goes up, which shouldn't (at least not as much) if only the CPU clock changed and not the custom chip clocks.

More specifically, using Frank Wille's memspeed v1.1 I get the following:

1)
50Mhz crystal, CPU@25Mhz:
----------------------------------------
ChipRAM: 3998 (read) 3990 (write) 1995 (copy)
FastRAM: 11427 (read) 6070 (write) 4006 (copy)
ROM: 10310 (read)
Note: default standard setup (25Mhz rated CPU running at 25Mhz)

2)
55Mhz crystal, CPU@55Mhz by using Macintosh "doubler" module:
----------------------------------------
ChipRAM: 3971 (read) 3943 (write) 1989 (copy)
FastRAM: 11493 (read) 6003 (write) 3913 (copy)
ROM: 10476 (read)
Note: 40Mhz rated CPU speed doubled. I *had* to use 55Mhz because it wouldn't "take" at 50Mhz. Stupid A3640 timing problems. MAPROM disabled.

3)
66Mhz crystal, CPU@33Mhz:
----------------------------------------
ChipRAM: 4941 (read) 4898 (write) 2461 (copy)
FastRAM: 15017 (read) 8072 (write) 5219 (copy)
ROM: 13565 (read)
Note: overclocked setup (25Mhz rated CPU running at 33Mhz)

4)
74Mhz crystal, CPU@37Mhz:
----------------------------------------
ChipRAM: 6880 (read) 6967 (write) 3484 (copy)
FastRAM: 16960 (read) 9070 (write) 5863 (copy)
ROM: 15252 (read)
Note: overclocked setup (40Mhz rated CPU running at 37Mhz)


General notes: all units in kB/s, 4000D-CR mobo without riser card, A3640 v3.1, Kickstart 3.1, Ramsey skip bit optimization off


Analysis:
Compare setup 1 and 2. With more than 2x the CPU speed the memory benchmarks remain the same. Now compare setup 2 and 3. In setup 3 the CPU is slower by 33/55=60% but the memory performance blows away the 55Mhz CPU setup because the custom chips/bus is running at 33Mhz vs. 55/2=27.5Mhz! Respectively for 2 vs. 4, only the difference is even greater.


Cheers!

@arnljot: yes, you do want to overclock this A3640 bitch ;-)

EDIT:
What I would like to do if I have time is to put a thermostat on Ramsey and FatGary (and possibly some of the other custom chips) and see what their thermal fingerprints are before and after the overclock. While I had the fan blowing on the CPU, I didn't notice them being hot at all, but some air was going over them as well (I had the blower on the side, covering the top and bottom of the A3640).
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

@da9000

Very interesting reading my friend, always good to see the results in print.
I'm going to have a good look at the schematics and try to piece this together but I do wish I had the schems on a single large sheet than small A4 bits & pieces, doesn't make the job easier :laugh:

Best Regards as always
TC :mrgreen:
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

I hear ya TC (about the schematics being all cut up like a puzzle). I've been meaning to print them out and paste them in one big blueprint but haven't done so yet :-( I should also look at the schematics but unfortunately I have to sleep soon coz of work and by the time work ends, well I don't have much time/energy to do much :-( So I'll count on you my friend :-)
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Kin Hell said:
I expect it should pull off the 060 CPU okay but I dont have a working socket to fit your 040 in other than a knackered CSMKIII that has a reworked and rebuilt socket array. Whilst any CPU in there does still get warm, it's an absolute sod of a job getting any CPU into the rebuilt socket due to the slightly irregular pin socket alignment. I did swap out 060's to take the CPU being dead out of the equation but it was a 2 hour event! :wooha:

If I pull the cooler off the 060, it leaves a sticky Thermal Pad on the Heatsink which I can not protect once removed. If I could drop the 040 into my dead CSMKII, I'd have to desolder the Voltage regulator Resistor on the CSMKIII in the home built socket in order to Juice the 040, thereby getting the Thermal pad warm in order to get it stuckk on for real. All these bloody new Icons & not a single "Flogging a Dead Horse" smiley anywhere! :stick:

I might be able to find a bit of plastic somewhere to press on for protection for shipping to Anthony, but it could all be in vain anyway. I really need to fire the
Click to expand...

Kinnie, I have absolutely NO idea what you're talking about up here, but I got one bit: something about an 040 socket. Let me know if you need one. I have. If you want to reword it in English this time (and not Kinnish :-p), then please do so. Perhaps I can help further.

Kin Hell said:
Beer time! :beer:

Charlie
Click to expand...

OK, now I understand what happened... :-D
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

da9000 said:
I hear ya TC (about the schematics being all cut up like a puzzle). I've been meaning to print them out and paste them in one big blueprint but haven't done so yet :-( I should also look at the schematics but unfortunately I have to sleep soon coz of work and by the time work ends, well I don't have much time/energy to do much :-( So I'll count on you my friend :-)
Click to expand...


Puzzle aint the word my friend, Nightmare :ROTFLOL2:

Do you know if you compare the A1200 & A4000 Schematics, even common signal lines are named differently between the two platforms :nuts:

I'll see what I can come up with.

Nighty Night
TC :wink:
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

@ TC @ da9000

Look at the X-Tals on the A4KD Mobo. 50Mhz & 28.xxxMhz

In the case of the A4KD Rev B (Rev D has different locations & orientation of X-Tals) the 50Mhz is where the 25Mhz 030 gets its speed from. The Black Rectangular Chip next to the 50Mhz X-Tal having 8 widely spaced legs is a Divide by 2 Counter, reducing the 50Mhz X-Tal to 25Mhz & is where the 030 CPU gets it's speed from with the INT/EXT jumpers set to INT. Don't forget, the 030 Cards in the A4KD also have their own Jumper settings for implementing use of either a PGA or PLCC type packaged FPU. These jumpers can be set in such a manner to either use the mobos 50Mhz X-Tal & divide by 2 counter to set the FPU @ the same speed of 25Mhz as the CPU on the 030 card, or.......you can fit an X-Tal of what ever frequency on the A3630 card to run the FPU @ a clock speed independant from the 030 CPU Speed.
J100 1-2 Use FPU in PLCC Socket
2-3 Use FPU in PGA Socket
J101 1-2 Use optional on-board oscillator at U103 for FPU Clock
2-3 Use CPU Clock as FPU Clock

The C= A3640 also Employs this arrangement of 50Mhz X-Tal & divide by 2 counter to derive it's default 25Mhz speed on the 040 CPU, but both the INT/EXT mobo jumpers need to be set to EXT in order to use a C= A3640 card. There is no option of jumpers on the A3640 Card for FPU's as the FPU is in the 040 Chip itself.

There is no need to go looking up this that & the other about overclocks affecting/effecting the Amiga's Chipset because the 28Mhz X-Tal is the one dealing with the Chipset speed & the Bus speed across the mobo. You will also find there are slightly different 28.xxxxMhz X-Tal speeds across the range of A3000, A4000 & A1200 for example...but this is just a brand thing & not required for the Bus or Chipset to operate. The Bus & Chipset has a range in which it can operate effectively.

RE going further into specific speeds crossed with whatever other factors, the above is basis for overclocking the CPU only & as to the CPU affecting/effecting bus speeds & ram speeds in the Amiga, I have to say a Firm no & appreciate TC already endorsed this fact.

As final proof in trying to explain these fundamental principles, take a CSMKIII card or a CSPPC card. There is facility on these cards to totally seperate the processor cards CPU & Bus Speeds & if you read my Overclocking CSPPC thread on AmiBay, I exploited this facility. So much so & to such an extent, I burnt my CSMKIII beyond repair as the Cards logic now appears to be "Toast". :blink:


Hope this helps, even if in Kinnie Language! :lol:

All the best chaps,

Charlie
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

The query I wanted to clear up in my mind was regarding RAMSEY being clocked by the chipset oscillator or by the CPU's as we know the ext Clock CPU does get fed back to the board via the CPU slot.

So it looks like the onboard RAM and Zorro III bandwidth is fixed then and runs asynchronous to the CPU.

TC :D
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

In regards to xtal on mb vs on cpu.

I've used 25mhz 030 and 50mhz 060 on my A4000. But never noticed that the mb runs hotter with the CS 060. So doesn't that confirm that the MB is asynchronous to the CPU card in that respect?
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

arnljot said:
In regards to xtal on mb vs on cpu.

I've used 25mhz 030 and 50mhz 060 on my A4000. But never noticed that the mb runs hotter with the CS 060. So doesn't that confirm that the MB is asynchronous to the CPU card in that respect?
Click to expand...

Exact-a-mundo! :mrgreen:

However, a Warp engined 040 @ 40Mhz will dissipate more heat than an 060 @ 50Mhz & Air flow in an A4KD is not as good as perhaps an A4KT is. Consequently, a 40Mhz 040 in a desktop Amiga without added airflow could theoretically warm up the entire environment within the case, thereby making the Amiga Chipset run slightly hotter than normal. :wink:

Kin
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Arnljot,

I recieved an Email from Anthony Hoffman yesterday. Your 40Mhz 040 CPU, 66Mhz X-Tal & heatsink/Fan combo are now in his hands, waiting for your stuff to arrive from Norway! :beer:

All the best m8y,

Kin
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

:yourock:

Noted, I've behaved accordingly as required of me by the amiBay regulations :run:
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

So I see! :lol:

Kin
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Thread closed, then. :wink:

[EDIT]As Arnjot requested, thread is re-opened.
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Thank you, I need to keep this open. I want to inform you all when I receive the card in return so that y'all know how it ends ;-)
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Arnljot,

I posted your Rev D Cr A4KD Mobo this morning. You should have it in 3 days. Have sent you an Email with all the Deets.

Let me know when you have her!

Best for now,

Charlie
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

I received a mail from Anthony yesterday, my A3640 cannot work at speeds above 25mhz, not 30, 33 or 40. 25 thats it.

But he has upgraded it to 3.2, so there is some concellation in that.

So my boing balls have shrunk, and are normal size now.
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Unlucky m8y! :|

Kin
 
Re: Q:Commodore A3640 tinkering - Watch out I have boing balls!

Report back from Anthony, my card could not handle speeds above 25mhz. So that's end of story I guess?

My boing balls are normal size it seems.
 
You must log in or register to reply here.
Back
Top Bottom