713 replies to this topic

[Claws]

Sounds like someone was in for the Renault job and was pipped by some fresh Uni guys

Edited by Claws, 04 March 2014 - 09:18 AM.

[JG]

Well most of it make no sense, and then you get to the terms like 'epic fail' and 'utter crap' which if you haven't discredited the author already (by around line 2) should serve a a pretty good warning.

[Claws]

I discredited him at "Magna cum laude" I don't know what that means and I'm sure as hell not gonna type that in to a search engine

[techieboy]

as a result the green monstrosity from Leafield would come second to most of my furniture in the wind tunnel.

[slindborg]

That is the most eloquent (spelling) I have ever read. Its very accurate and very applicable to many industries and situations

[Gedi]

What a load of nonsense. I thought it was an accidental post, until I actually read it. Well, I got half way thought and stopped, the author clearly doesn't know what he's talking about  

Back in 1994 I installed Linux on my computer for the first time. <snip> So I spent two weeks designing and programming device drivers for my exotic hardware.

Bullshit. Complex drivers takes months / years to develop. If he's installed linux for the first time, he's definitely not writing drivers in 2 weeks.....  

List BunchOBolts = new List (); List BunchONuts = new List (); Nuts nut=BunchONuts.begin(); for (Bolt element in BunchOBolts) { element.ConnectTo(nut); nut=BunchONuts.next(); } Voila! Program done in 2 minutes: Instant gratification. Now heres the problem. They have no control over whether or not the Nuts and Bolts are created properly and with a minimum amount of resources. And more importantly, they rely on the runtime system to give the resources used by the temporarily created Nuts and Bolts back to the system when they are no longer used.

More bullshit. C# may hold you hand if you tell it not too (which is how 99% of people will and should use it), but you can have complete control over it if you know what you're doing. This guy clearly doesn't..  

MOV BX,AX SHL AX,2 SHL BX,4 ADD AX,BX SHL AX,2

This is 16 bit assembler, it hasn't been used since the 90's unless someone it trying to access 16 bits of a 32/64 bit register. He clearly copied this from somewhere. (it actually just bit-shifts a number by 2 and 4 and adds them back together. Nothing to do with his previous example) I got bored shortly after this and stopped reading. I'd bet my house on it that Renault haven't hired C# programmers fresh out of college to develop the software that runs their F1 cars....

Edited by Gedi, 04 March 2014 - 09:39 AM.

[siztenboots]

I learnt assembler on a 6502 , which had no multiply operand, just one acc register and two index registers with distributive properties of multiplication, you can, for example, rearrange the problem of: 12*12=144 to (4+8)*12=144 to (12*4)+(12*8)=144 hence two SHL on one register , 4 SHL on the other and add , is a multiply by 12 now do do that with just one 8 bit register

[slindborg]

I got bored shortly after this and stopped reading. I'd bet my house on it that Renault haven't hired C# programmers fresh out of college to develop the software that runs their F1 cars....

 

I wouldnt be sooo sure.....

[simonlpearce]

And here was me expecting some F1 related stuff. All the above is like reading Chinese to me...

[siztenboots]

MUL AX * 80 , be the equivalent of multiply AX * 64 + BX * 16 SHL AX, 6 SHL BX, 4 ADD AX, BX

[G-Bob]

Don't they just wang the boost slider to the right until it was all the way up and that's your car at full power?

 

 

#AntiSlindborgTuningMethodology

[slindborg]

Don't they just wang the boost slider to the right until it was all the way up and that's your car at full power?

 

 

#AntiSlindborgTuningMethodology

stage -1

[Zoobeef]

Aaaaaaaaand youve lost me again.

[Claws]

Don't they just wang the boost slider to the right until it was all the way up and that's your car at full power?     #AntiSlindborgTuningMethodology

stage -1

Needs a dump valve innit

[Scuffers]

I learnt assembler on a 6502 , which had no multiply operand, just one acc register and two index registers with distributive properties of multiplication, you can, for example, rearrange the problem of: 12*12=144 to (4+8)*12=144 to (12*4)+(12*8)=144 hence two SHL on one register , 4 SHL on the other and add , is a multiply by 12 now do do that with just one 8 bit register

Snap!

 

6502 followed by 8085/Z80

 

What a load of nonsense. I thought it was an accidental post, until I actually read it. Well, I got half way thought and stopped, the author clearly doesn't know what he's talking about  

Back in 1994 I installed Linux on my computer for the first time. <snip> So I spent two weeks designing and programming device drivers for my exotic hardware.

Bullshit. Complex drivers takes months / years to develop. If he's installed linux for the first time, he's definitely not writing drivers in 2 weeks..... 

About that time I did much the same, running Linux on PC's and writing drivers for weird sh*t (in my case, 16 port Digi-boards for something not unrelated to this thread)

 

if you knew what you were doing, and understood machine code, then yes, you could write a driver in days, not weeks, hell, I was never 'good' as a programmer (nothing like geeky enough), but if I could manage it, it's hardly rocket science, and yes, I agree with him, modern software is a joke on so many levels, if they had to write stuff that worked on a Z80 at 4Mhz with 16K of ram, let's see how they get on.

[Gedi]

Writing drivers isn't about being able to understand machine code, anyone can learn that, it's about having a deep understanding of the architecture of the operating system you're writing for.   In the Windows world, I'd expect it to take an already experienced Windows programmer at least a year of learning/reading about windows internals before being able to write a proficient driver, and at least a further 5 years to be good enough to call yourself a decent kernel programmer.  

if you knew what you were doing, and understood machine code, then yes, you could write a driver in days, not weeks,

Absolute codswallop. I'm currently writing an encryption driver for NT6, I've been on it for 3 months so far, and I expect it to take a further 6 months before it's solid. I might not know much about cars, but I've been working with the Windows kernel for about 15 years now so I know this

Edited by Gedi, 04 March 2014 - 08:15 PM.

[turbobob]

Fortran ftw

[Scuffers]

Writing drivers isn't about being able to understand machine code, anyone can learn that, it's about having a deep understanding of the architecture of the operating system you're writing for.   In the Windows world, I'd expect it to take an already experienced Windows programmer at least a year of learning/reading about windows internals before being able to write a proficient driver, and at least a further 5 years to be good enough to call yourself a decent kernel programmer.  

if you knew what you were doing, and understood machine code, then yes, you could write a driver in days, not weeks,

Absolute codswallop. I'm currently writing an encryption driver for NT6, I've been on it for 3 months so far, and I expect it to take a further 6 months before it's solid. I might not know much about cars, but I've been working with the Windows kernel for about 15 years now so I know this

Who said anything about windows?

[turbobob]

Is it me or is this thread going a little off topic?

[JG]

It's you