Slashdot Mirror

And the award for people who need to learn to use google goes to (drum roll) TubeSteak.

BSAFE is a product by the RSA Software corporation that provides cryptographic functionality to developers. LibTomCrypt is my crypto library which I give out for free [under public domain] at http://libtomcrypt.org./ SCEA is the Sony Computer Entertain of America Corporation.

Tom

They wanted to get their Netburst cores into the DP world as quickly as possible.

Where AMD uses the HT bus for their 757 and 939/940 parts Intel was still using the good ole 64-bit FSB of yesteryear.

Most of what Intel does nowadays in the processor world is entirely market driven. The Netburst is a good example. High clock rate, low efficiency processor. Sounds good on paper but works poorly in practice. The EMT64 extensions are another example. A lot of code on the P4 in 32-bit mode takes roughly the same number of cycles on the 64-bit P4s with the notable exception being 64-bit math [e.g. additions and multiplies].

For example, most block ciphers are the same speed on both the 540J and 820D [in terms of clock cycles]. I think partially because they're just using rename registers for the additional GPRs. But compare the AthlonXP to the Athlon64 and there is a huge difference. The Athlon64 is an improvement over the 32-bit cousin. They didn't just slap 64-bits on the core they actually made it better.

I refer to my nice chart again

Operations per second at doing RSA-1024 decrypt

AMD64 = 2.2Ghz
AMD32 = 1.8Ghz
P4 = 3.2Ghz
Nocona = 2.8Ghz

At the 32-bit side of things the AMD32 can match or beat the P4 even though it's slower by 1.4Ghz. At the 64-bit side there simply is no comparison. I mean the dual-core RSA on the Nocona can't even match the SINGLE-CORE RSA on the Athlon64.

How pathetic is that?

Ever since the 64 came out Intel has basically been a poser in the CPU world. The only really proud achievements [outside the pure sciences they do in the background] are the ARM and P6 core designs...

Tom

Bad form replying to my own post [hence the AC]

ops per second.

FWIW, the AMD64 is at 2.2Ghz, AMD32 [Athlon-XP-M] @1.8Ghz, P4 @3.2Ghz, Nocona @2.8Ghz.

So yes, a 3.2Ghz P4 Prescott gets roughly the same number of RSA/sec as a 1.8Ghz Athlon-XP ... :-) There is a slight power consumption difference [one of them tops out at 110W the other 45W ... I'll let you do the math].

Tom

Why do they always do gaming as the benchmark? It's a server processor!!! Do some crypto!

Check this out image where "nocona" is a Pentium 820D [dual-core 64-bit P4].

Those are cycle counts for RSA-x private key operations [with padding] on various processors.

TFM == tomsfastmath
LTM == libtommath
DC == dual-core [two threaded] tomsfastmath :-) Shameless plug but also good numbers when doing RSA work I guess.

Tom

Just add a bit...

The PPC [G3] is about as efficient for bignum math as the AMD Athlon-XP and the P4 ALU. See this chart for instance.

I think the G4 maybe be slightly better [the instruction set remains the same w.r.t. bignum math] but still same ballpark.

The problem with that design really is that while it has a good RISC ISA and lots of registers it's simply not meant for math. You have to execute two 4-6 cycle multiplies to get one 32x32=>64 product whereas other cpus can get the full product in 6 cycles AT A HIGHER FREQUENCY.

Where the PPC is nice though... is in embedded work. It's not quite as efficient as the ARM but can be clocked higher. So it's a good tradeoff between something like an ARM which gets enough MIPS and way low power and an Athlon which gets very high MIPS but takes a lot of power [by comparison].

All in all though I think the PPC can hold it's own against the x86 offerings. It just doesn't scale as nicely.

I don't know about the G5 though [I don't have access to one]...

Tom

I dunno, college served me better than I guess

- Met some peeps who I'm still friends with

- Gave me something todo between the highly structured high school and consistently demanding full-time work

- Learned a thing or two I probably wouldn't have learned [by this point] on my own.

- Gave me some time to practice my craft when the living was cheap (free rent from parents ;-) ), etc....

Tom

That's not always true. Some languages do have idiosyncrasies that are nasty to overcome. Ruby for instance is nothing like C. Same thing with Python.

So if you put on your resume
- Mastery of C
- familiar with Java, Perl, C++

I wouldn't expect them to ask you to write a Python webserver or something "to see how you think" during an interview.

RiM [the blackberry makers] are big on quizzes too. I posted for a job online and they made me solve some "number of shapes with N sides" puzzle. I was 99% to an answer so they were a bit annoyed but decided to interview anyways. During the interview they gave me another question [about how to draw some form of pascal triangle].

While we were discussing the question I basically asked him "does solving puzzles make me a great employee?" to which he replied "it's good to know how you think" or some such.

See my background is crypto and math. I know how to multiply fast, how to exponentiate fast, etc. But because I took the full 3 minutes to write the algo [for the triangle thingy] he seemed not pleased as if I wasn't smart enough for the job.

So I then asked him to give me an algorithm to multiply 1024-bit numbers quickly [less than n^2 time] to put him in his place. He said he didn't know how and was mystified by the question. I told him "oh, well because YOU can't solve that puzzle you're a bad employee".

At which point the interview was over.

To me interviews are stressful. Giving stupid questions is retarded. Wanna see what I'm capable of? Check out the 100,368 lines of C, TeX and Perl that make up my public domain portfolio [e.g. libtom projects] that is used by corporations around the globe.

But somehow that doesn't count.

During an interview at entrust [a crypto company] they didn't seem to care about LibTomCrypt and more so if I could debug a Java program they printed out [it had a race condition]. I never said I was proficient at Java [my resume lists it but all my projects are in C].

Mostly I think interviewers don't do proper background checks. They [entrust] never downloaded a copy of LibTomCrypt to look at it. The questions were not suited to my resume [they called me first!] and I left feeling stupid because I couldn't answer the questions properly. ... just my experience...

Tom

Someone port Libtomcrypt to it!
It compiles in ARM CPU's already, I believe.
Crypto on your calculator!

Forget about libgcrypt. It's documentation is poor anyway. Instead you should use LibTomCrypt, which is free as in true freedom (public domain), and has compiled on every system I've tried it on (including Cygwin).

(No I'm not the author... just a satisfied user)

You could use the GMP library, but it's GNU code so you might want to be careful about licensing.

For a real free bignum library, try LibTomMath. It was written by a guy named Tom.

That isn't always true though with open source. Look at GNUPG's /cipher directory? It's been stagnant for nearly a year and the code quality is simply pathetic. Sure it "works" but if OSS is supposed to furnish the best quality code then it ought to at least allow for improvements.

I've sent patches to the GNUPG folks and they flat out refused to even look at them let alone merge them in.

Say what you want about me or the quality of my code but if you actually look at the patches I wrote you can appreciate the things I fixed up for them. Nothing earth-shattering but still a clean-up nonetheless.

My point is YMMV with OSS just like proprietary software. Some people just don't like being told they're wrong.

Tom

What's the problem? Just use gnupg [see this memo for more info].

Tom St. Denis has a nice crypto library at libtomcrypt.org. I like it, anyway.. it's worth a look :)

I don't like the GPL if that's what you mean... but I do happen to write some public domain software as well as a public domain textbook (included in the LibTomMath package).

Tom

I don't like the GPL if that's what you mean... but I do happen to write some public domain software as well as a public domain textbook (included in the LibTomMath package).

Tom

I don't like the GPL if that's what you mean... but I do happen to write some public domain software as well as a public domain textbook (included in the LibTomMath package).

Tom

I don't like the GPL if that's what you mean... but I do happen to write some public domain software as well as a public domain textbook (included in the LibTomMath package).

Tom

libtomcrypt has an implementation of ECC as well as RSA and DSA/DSS as well as a bunch of hashing algorithms and symmetric ciphers.. And it's free.

How did I spam my own shite it isn't like I am trying to draw attention to myself.

See, now **that** is spamming.

How did I spam my own shite it isn't like I am trying to draw attention to myself.

See, now **that** is spamming.

How did I spam my own shite it isn't like I am trying to draw attention to myself.

See, now **that** is spamming.