cabr07@vaxa.strath.ac.uk (05/28/91)
A report in the *April* issue of RISC User made reference to a new reduced spec for ARM3s. Seemingly they have been reduced to a maximum speed of 25 MHz to increase the chip yield. I have read similar reports from other sources so I guess this is true. The article then goes on to say that the maximum clock period from MEMC has been reduced from 10000 ns to 250 ns rendering the chips useless in Arcs. Can anyone from ARM Ltd confirm this second part before I part with my cash for a chip I couldn't use? ---You can never be sure with these April issues--- Gordon. -- -------------------------------------------------------------------------------- |G.N.Sinclair | JANET: G.N.Sinclair@uk.ac.strath.vaxa | |Physics & Applied Phys Dept| BITNET: G.N.Sinclair%vaxa.strath.ac.uk@UKACRL | |Strathclyde University | UUCP: G.N.Sinclair%vaxa.strath.ac.uk@ukc.uucp | |107 Rottenrow |Internet:G.N.Sinclair%vaxa.strath.ac.uk@nsfnet-relay.ac.uk| |Glasgow G4 0NG, Scotland | | --------------------------------------------------------------------------------
gtoal@castle.ed.ac.uk (G Toal) (05/29/91)
In article <1991May28.160336.11750@vaxa.strath.ac.uk> cabr07@vaxa.strath.ac.uk writes: >A report in the *April* issue of RISC User made reference to a new reduced >spec for ARM3s. Seemingly they have been reduced to a maximum speed of >25 MHz to increase the chip yield. I have read similar reports from other >sources so I guess this is true. I suspect this is true. I bought a 540 and was surprised that it didn't run a particular program at quite the speed that Les Currell was claiming for his home-made Arm 3 upgrade. I suspect the 540's as sold aren't running at maximum speed; I haven't got the dhrystone code out yet to check and I haven't been bothered by it enough to open it up and look at the crystal. If anyone really wants me to I will. If it turns out that the machine is rated at less than the speed claimed in the advert I looked at just before I bought it, I wonder what I should do? ;-) [relax Acorn - it's fast enough for me :)] Graham
gilbertd@p4.cs.man.ac.uk (Dave Gilbert) (05/29/91)
In <1991May28.160336.11750@vaxa.strath.ac.uk> cabr07@vaxa.strath.ac.uk writes: >A report in the *April* issue of RISC User made reference to a new reduced >spec for ARM3s. In this months Risc User I seem to remember a little note from ARM or VLSI or one of the associated people syaing that although they had down spec'ed it they had given guarantees on its suitability for present Arm machines. (I think). Dave -- ------------------------------------------------------------------------------- - Dave Gilbert - gilbertd@p4.cs.man.ac.uk - The MTBF of a piece of equipment - - G7FHJ@GB7NWP - is inversly proportional to its - ------------------------------------------- importance -
asmith@acorn.co.uk (Andy Smith) (05/29/91)
In article <10615@castle.ed.ac.uk> gtoal@castle.ed.ac.uk (G Toal) writes: >I suspect this is true. I bought a 540 and was surprised that it didn't >run a particular program at quite the speed that Les Currell was claiming >for his home-made Arm 3 upgrade. I suspect the 540's as sold aren't >running at maximum speed; ?? This is strange. Tests we have run on 440/1's with Aleph one ARM 3 cards and 540's, show the 540 to be faster. This is mainly due to the faster memory speed in the 540. I havn't seen Les Currell's claims, but I doubt a 400 machine could match a 500 machine (though I'm open to proof). I don't know what the official word is, but I wouldn't buy an ARM 3 upgrade spec'd at over 26Mhz now, as VLSI don't sell them. These tests were done with a very complexed draw file, and are nothing to do with Acorn's official testing of machines. >Graham Andy