....http://www.schoolmart.com/ has SE for $121

dude, like its a cow

Mine's in the mail! :) Can't wait. Looks cool and sounds cool. Selling my old one ordinary 83+. Comes with everything. Anyone intrested?

I know this is off topic, but this question still goes unanswered despite many posts on previous message boards.

"How can SFII be run on an HW2 89?"
-You cannot put AMS1.0 on an HW2
-This game needs AMS1.0 to work correctly (or so I'm told by its authors)

I have had people tell me that they have gotten it to run, but they havn't explained how.

-If you know how to get SFII to run on an HW2 89, please tell the rest of us wanting to play it how you have done so. I'm sure that the 89 crowd will glorify you (if your method works).

Zach Hobson

Some time ago Flavien Racine (sp?) Dissassembled patched and reassembled the original SFII to work with any hw/ams. He than asked the authors if he could distribute this version from his website. The authors claimed to still be working on it, and didn't approve, so Flavien didn't distribute it.

If someone is claiming he is runing SFII on a hw2 he is either:

* Runing a copy of Flaviens patched version that got out somehow.

* Runing a similar patched version of some other source.

* Just lying.

My best bet how to get it to work is to contact somebody who claims to have it working, and ask him how...

By the way, there is a new fighting game for ti89 released called 'Streets of Rage'. Check it out on www.ti-fr.org . Quite impressive... The instructions are in french but not to hard to decipher with a translation web site and some patience. I don't know if it works with the hw2, but it works with my hw1, ams2.05.

Thank you for the help. I hope that Peter Herringer will release SFII re-coded soon...

If anyone else knows how to get SFII to run on an HW2, don't hesitate to respond! (please reply)

Zach Hobson

can you give me some help on how to get to the program files for streets of rage and the other programs with screenshots on ti-fr? I can't seem to find the link :(

I find it sad that you couldn't find the link.

Go to www.ti-fr.org and scroll down the main page. You will see an article with a single picture of SOR3, and some text. Within the text there is a yellow link entitled "Streets Of Rage". Try clicking that...

If that is to hard, here is the direct link:
www.ti-fr.org/~prog/temp/sor3.zip

You probably don't need to be playing games in class if you can't even find the link to the file you want.

--TI-89 HW2 Street Fighter II Compatability--
¤I'm still looking for anyone who knows how to get this to work (Flavian's version is great [and runs], but is ridden with bugs)
¤If anyone knows whether or not Peter Herringer (sp?) is going to re-compile SFII for HW2/AMS2.0x, please let me know!

Zach Hobson

sorry, your post just makes me laugh :)
My bad, i was looking at the Streets of Rage / NonOS / Thunder / Half-life screenshots. According to altavista, the english translation says that those four programs aren't available yet due to bug issues. I assumed that you had some other link, seeing as the page itself most recently said that the program isn't available. I didn't bother to read down further, which I guess is my fault for being an imbecile. Now, as to the personal attacks, completely unneeded:

"You probably don't need to be playing games in class if you can't even find the link to the file you want."

"If that is to hard (sic), here is the direct link:
www.ti-fr.org/~prog/temp/sor3.zip"

Yeah, I admit it, i'm a frickin' moron. I really don't want to brag, but you've insulted me. I got a 1600 on my SATs, 240 on my PSATs, all 4's in the highest-level classes my school offers, GPA well over 4.2, #1 in class currently, scored high on some national math test, etc. Call me crazy, but I think i've got a pretty good idea when to be playing games, and it's only during non-classwork time.

I thank you for your help, but I find it sad that you feel insecure enough to have to personally attack someone who makes an innocent mistake.

I'm sorry I insulted you, I assumed that you were a middle-schooler who had glasses too thick to see the link. No personal offense intended.

As for your comments about me being insecure--I literally laughed to myself when I read that. I myself look down upon others who have the quality of being insecure with themselves. I was not trying to raise my self-esteem. I (not to brag) am a very humble person and am not afraid to be different. For instance: I wear Chuck Taylor All Stars and am an old-school Metallica, Megadeth, and even Iron Maiden fan. I realize that you were most likely retaliating for the comments I made at you. No offense taken (since I know that quality doesn't apply to me anyways). As for your perfect SAT score, congratulations (as well as your GPA, etc.). My current GPA is about a 4.3333, so please don't view my improper use of the form of "to" too badly.

I hope you feel differently about me now, and I hope you have fun with Streets Of Rage 3. Next time, just try scrolling down the page. :)

Concerning playing games in class: I havn't paid attention for over 10 seconds in my math teacher's class this semester, and I have a 97.5%. Some of us <I>can</I> play games and get good grades too (with the added help of a TI-89 [I understand the concepts we are being taught, the calculator just makes work/tests mush easier]) ;)

Zach Hobson

see, that's the problem with trying to get personal around here :) anyone that actually signs up for an account and posts on these things is obviously a closet nerd, so we're generally pretty bright, and don't have to pay too much attention (although I do sit up front in all of my calculator-equipped classes now, hmm... guess they're catching on - thank god for f5) Sorry for getting upset, today's been a bad day. Thanks for your help... too bad I can't get it to work HW2, AMS 2.05, etc. I'll wait a bit, it'll work later - i'm plenty content with this CTC2 demo.

apologies,
Jon

Phew! I'm glad this whole issue is settled. I was kinda in a bad/criticizing mood also (I have the bad habit of staying up until 1:00 A.M.ish).

As for the closet nerd thing--yeah, I'd think that most people who even visit TICalc are pretty nerdy. Lukily for me, I look like an 80's metalhead (and I am one) and hide all my nerdy qualities while at school (just so I can hang out with the gals ;) ). Another thing on my side is that I'm a 7th year football player (4th year varsity) that lifts/conditions every day and weighs 195 lbs.--and I don't have glasses. All these qualities help to hide the fact that the majority of my time at home is spent basking in front of my PS2 and TV. lol.


Zach Hobson

PS2!?!? Lucky. 1:00? Is that all? I am usually up until 4, watching TV, writing books, and sleeping. I unfortunately was classified as a nerd early in my life. but no one bothers people who are home schooled.

Ignore this one

PS2!?!? Lucky. 1:00? Is that all? I am usually up until 4, watching TV, writing books, and programming. I unfortunately was classified as a nerd early in my life. but no one bothers people who are home schooled.

Yes, PS2, and sometimes later than 1:00 :).

I'm also a shredder on guitar :p.

Long live the nerds.

Zach Hobson

In the news at calc.org, it is mentioned that a ROM Call is used to switch the calculator from normal mode to overclocked mode. As is TI's bad habit, does this mean the ROMCALL table is somewhat rearranged? Because of the switches, features that use the extra power would change size in the ROM. This could change the entire thing--is there anything I am missing?

As for programming, it looks as if the Ion/Mirage build will have to be modified to allow for entry of this command in processor-intensive programming.

One more thing to programmers who need more FLOPs: the Z80IS has faster execution times than the 68K; also, the processor is 1.5x the 89. Unless there are other factors, like the FSB or something, the new calculator will be more powerful, speaking strictly in FLOPs, than the 89.

Leading to my final point, it is disappointing to see that TI is slacking off 89/92 series support. The superior screen and programming capability of the 89, versus the raw power if 83+SE, is too confusing to provide a true benchmark. Whatever happened to top-of-the-line?

S11001001
That's right, I don't use C :]

MHZ, speed, power, these are different to the Z80 and 68k.

15 mhz for a Z80 does NOT equal 15 mhz for a 68k. If you think that ANY Z80 calc is superior to the 89, do us all a favor and get CAT scan.

The Zilog z80 is an 8 bit microprocessor, while the Motorola 68000 is a 16 bit microprocessor. Also, the hw2 89/92+ 68000 chip runs at 12MHz, not 10MHz. Unless TI used an incredibly inefficient compiler on the 68000 AMSes, the z80 should have no chance of beating the 68000, possibly unless it is overclocked to something like 60MHz, but that would almost certainly overheat the chip.

Side query: does anyone know if the 89 can be overclocked safely?

Yes you can overclock a Ti-89. I have overclocked my HW1 to 18.3MHz (about 2 times faster). There are several sites out there with instructions. I do not reccomend the "variable capacitor" method. I do reccomened the speed switch. Good luck ^_~

As far as the speeds of processors go: the 68000 is faster!

Thanks for the info. And we all (should) know the truth about the TI calc microchips already, but you're right.

FLOPS (floating-point operations per second) are not purely determined by clock speed. Chip design is a significant factor in the FLOPS of a chip at a given clock speed. For example, according to GIMPS' benchmarks (link above), the AMD Athlon 900MHz and faster outperforms the Intel Pentium 4 1500MHz (Fast Fourier Transforms are essentially nontrivial floating point calculations). While both chips use superscalar architectures, the Athlon has 3 independent floating point pipelines, while, AFAIK, the P4 has only one floating point pipeline. And both of these chips are 32-bit, so the difference between the z80 and 68000, especially if they were run clock-for-clock, would be even sharper.

The SE's FLOPS are a flop compared to the 89's, I am sure.
Sorry, I had to say that. :) Top of the line=89.

I already know the 89 is superior. Why do you think I write ASM for 89/92+ instead of 83? But the point is, the Z80 is far RISCier than the 68k. I am taking reciprocity into account when I see the movement capabilities of the 68k, which are far superior to those of the z80.

Also, I am taking into account that FLOPs are faster in 83 BASIC than 89 BASIC. Don't believe it? If you have both, get my BASIC Prime Factor v1.0 for both calculators and plug in a rather large number.

Finally, FL pipes and L2 do not apply to z80, since the FSB is plenty fast enough for the processor. Speaking of SETI@home, what do you think of putting it on the TI?

It is certainly not generally true that Z80 instructions are "faster" than 68K ones in actual execution, as can be seen if you look at charts of instruction times. On both processors, it is true that the simplest instructions take 4 clock cycles each. However, on the Z80 this will only work for byte-size calculations; to add two 16-bit values on a Z80 takes 11 clock cycles, while the 68000 can do the same thing in 4. The situation would be even worse if you wanted to AND, OR, or XOR two 16-bit values, as you'd need several instructions for this. Additionally, that the Z80 has fewer registers (something definitely uncharecteristic of fast RISC machines), and restricts which registers operations can be done with, greatly increasing the amount of data movement that is likely to be needed. And even though the 68000's multiply and divide instructions are slow, they are still far faster than the code needed to do the same thing on a Z80. Combine these things with the more flexible addressing modes of the 68000 which allow one instruction to do in a fairly small number of clock cycles what would take many instructions and many more clock cycles (not to mention sometimes temporary storage) on the Z80, and the Z80 no longer seems even close to matching a 68000 even if it does have 1.5 times the clock rate.

As far as "real" tests of speed, I don't know of any. I have found some results of the Dhrystone integer benchmark (not a 100% ideal measure of speed in itself, of course) on the web, and depending on which run of it you are using (with speed varying according to compiler optimization, and other things) the 68000 is between 2.44 and 5.77 times as fast as a Z80 at the same clock rate. In my brief search, though, I didn't find any Whetstone or other floating point results on both.

The speed of a basic factoring program is hardly a complete comparison of architectures, as I would expect the interpreters to work quite differently on the two calculators, and when a basic program is running the processor time is certainly not all used on calculations. On either calculator, the number of FLOPS obtained by such testing will be well under the actual number of floating point calculations the processor can do, not only because of the interpreter's overhead, but also because the calculator's use a BCD format for floating point which lowers performance compared to binary floating point.

And on a final note, it is a well-established fact that the TI-89's ROM was made using an inefficient compiler. To quote the great Zeljko Juric, "It seems that whole AMS is written using a non-optimizing compiler ... Some AMS routines are so badly written that I sometimes wonder whether any C compiler can be so non-optimal... "

How good do you think the AMS would be if it was made using TIGCC?

If TIGCC is an optimizing compiler (I assume it is, but honestly don't know), it would improve at least somewhat, although almost certainly not as fast as it would be if the AMS was rewritten in hand-hacked assembly or machine code.

Actually, that was a link to the "Great Internet Mersenne Prime Search", not SETI@home. I used to run the latter, but my parents for some reason wouldn't let me continue doing so. I recently rediscovered GIMPS and have been using it for about a month now.

When comparing calculators using basic programs, remember that each interpreter is different, and there are also several other factors in FLOPS computations. Also, I don't know what methods your programs use, but FLOPS are irrelevant to integer calculations, with the exception of FFTs and IFFTs.

Unfortunatly, SETI@home will not release any of its source code to protect its scientific integrity. So it would be *very* hard to port to a calc.