From: Annie on 12 Sep 2005 17:44 On 2005-09-12 Randy Hyde said: > Betov wrote: > > > Shut up, > > Telling me to shut up isn't going to do the trick. You've been told > many times how to get me to leave you and your product alone. It's > quite simple, really. All you have to do is stop insulting HLA > users, HLA, MASM, and other products and people around here. > Concentrate on your own product rather than everyone else's. Start > behaving in a civil manner. And I will quickly forget that you and > your product even exist. _____ Holy NetKKKop, 'Professor!' ((( `\ Hehe! _ _`\ ) (^ ) ) ~-( ) _'((,,,))) ,-' \_/ `\ ( , | `-.-'`-.-'/|_| \ / | | =()=: / ,' aa
From: randyhyde@earthlink.net on 12 Sep 2005 19:28 Paul Dunn wrote: > Betov wrote: > > [Some stuff that's not really relevant] > > Betov, did you not read my last question? I'm interested in your definition > of "emulation", so I understand what you mean by it. I don't think it's the > same definition that I use. > > Can you explain? I think I can explain it for you. Given Rene's recent attempts at explaining his algorithms, I doubt his explanation will be satisfactory. However, note that my explanation is based on what I understand his explanations to say, I've not bothered wasting any time trying to read his source code (see Alex's discussions for the problem with that approach). What Rene is doing is making multiple passes over some encoded representation of his source code rather than multiple passes over the source code itself. He believes this technique to be unique (because he thought it up all by himself, I suppose). I guess that he's unaware that this is how modern compilers all work. Bottom line, though, is that he does make multiple passes (the correct phrase is "multi-phase") over each of the conditional jump instructions in order to shorten them down. His code is a bit immature and doesn't do as good a job as assemblers such as Gas or MASM (which I've compared his code to, I suspect FASM does a better job too, though I've not tried this); worse, he only handles conditional jumps and doesn't optimize unconditional jumps or other instructions whose size is affected by the distance between labels in the code. Fixing the problem with conditional jumps that I've described in an earlier post is probably an easy fix. It probably won't take Rene more than a couple of hours (if he hasn't fixed this already). OTOH, allowing for other instructions than condition jumps sounds like a whopper of a design defect, and unless he's lucky this fix could take a lot more effort. Cheers, Randy Hyde
From: Betov on 13 Sep 2005 04:30 "anonymous" <wrigley(a)tuknet.dk> ýcrivait news:89e4e$4325f70a$3e3d8699$17300(a)news.arrownet.dk: > Doesn't all that anger cause you constipation? If you re-thought your > vocabulary you could save a lot of effort, and you and anyone reading > what you write would inevitably become happier in some/every way. For > instance, the above could have been expressed something like this: > > "I generally disagree with you, so please do not make negative > comments to what I write, as this will not make me change the features > anyway. > > "As long as it can be proven that RosAsm is without those faults, that > you claim it has, I shall not fix them, afterall you can't fix > non-existing bugs. > > "Feel free to do other more constructive things with your time" > > This was merely a sample, naturally you would want to add your own > unique style to it. Have a go! Once you get started posting happier > posts, you won't be able to stop. :)) In real life, i would much prefer knocking on the face of such an ####, with my hands of old home maker. This would be way faster, and would make me way more happier. :)) Betov. < http://rosasm.org >
From: Richard Cooper on 18 Sep 2005 13:41 On Sun, 18 Sep 2005 11:15:09 -0400, Betov <betov(a)free.fr> wrote: > There is only a nobody attacking RosAsm. Perhaps, but reading Randy's post, I found many reasons I wouldn't want to use RosAsm. Now if he'd just write a similar post for all of the other assemblers then it'd be a lot easier for me to decide which ones I'd like to have a closer look at.
From: Guga on 18 Sep 2005 15:34
And this is a problem, because ....... ? Randalk said so, i presume ;) I strongly advice you to stick in HLA then. If you take Randall´s advices serious, you really should use his product, either it is not considered an assembler. (not only by me) Best Regards, Guga |