A collection of Atari BASIC benchmarks

[Mazzspeed]

On 6/9/2021 at 11:26 PM, Stephen said:

 

Just leaving these here for you again since this is the only point some of us are trying to get through your thick head.

The above is post #161 for further context.

9 hours ago, Stephen said:

Oh wow - look at that.  I guess I'm not the buffoon with the problem, and I was correct in calling out the DMA issue.  Thanks for the validation.

 

I put that in bold and big text so it can show up easily, even if we don't have a theater scaler in the way of our pure high bandwidth signal rendering chain.

Did you see these posts? I can go through your posting history from the moment you joined AA, and the ignorance is immediately evident. As highlighted by the above posts, I'm not the only one that's noticed.

 

It's time this thread got back on topic, no one's interested in your pointless ranting. You results are interesting, but unfortunately invalid as highlighted by more people than just myself.

Edited June 12, 2021 by Mazzspeed

[Faicuai]

1 hour ago, drpeter said:

FCC regulations of the time were responsible for a lot of the costly build design decisions of the original 400/800 series.

Absolutely, and there is something else..

 

The design of the 400/800 arise from having Apple as their real competition, besides FCC regulations. And it shows in the 800 architecture, which in part resembles a bit the Apple-II's (oozing) open-spirit, with ease-of-use in mind, as well.

 

This is why I thought of pertinent including the Bit3 (80-col. output) as a valid operating environment for the tests, which will be more CPU-biased (no need of Antic at all, also supported in OS/b)

 

[Faicuai]

58 minutes ago, Mazzspeed said:

Did you see these posts?

I could not... just too many C64-adds to wade through...

 

But I was lucky enough to see this one, coming after those...

 

 

For the purpose of this thread's topic (A8 Basic repo.), we'll pick it up from here...

 

You can now go back to your kiddie computer and your plushy bear. ??

[Mazzspeed]

13 minutes ago, Faicuai said:

I could not... just too many C64-adds to wade through...

 

But I was lucky enough to see this one, coming after those...

 

 

For the purpose of this thread's topic (A8 Basic repo.), we'll pick it up from here...

I think we can pick it up from this point on. Your results, as interesting as they were, are invalid. They've been discussed far more than needed, and there's no reason to discuss them any further.

 

In fact, I see no reason to discuss the results of any benchmarks whatsoever as that was never the purpose of this thread from the onset.

[Faicuai]

33 minutes ago, Mazzspeed said:

I see no reason to discuss the results of any benchmarks whatsoever

What you see or don't see, what you think or don't (as a non-contributor of this Repo) is essentially irrelevant.

 

What really matters is what the creator of that Repo (Maury) truly wishes to accomplish with whatever gets loaded / shown there, and his motivations for doing so. Same case with Bench64's repo. 

 

That is what needs to be tracked (nothing else).

 

 

 

[Mazzspeed]

25 minutes ago, Faicuai said:

What you see or don't see, what you think or don't (as a non-contributor of this Repo) is essentially irrelevant.

 

What really matters is what the creator of that Repo (Maury) truly wishes to accomplish with whatever gets loaded / shown there, and his motivations for doing so. Same case with Bench64's repo. 

 

That is what needs to be tracked (nothing else).

 

 

 

Well perhaps you should have thought of that before derailing his entire thread in a vain attempt to show off before considering Bench64's benchmarking requirements - A repository you don't contribute to.

 

Pot calling kettle black.

Edited June 12, 2021 by Mazzspeed

[Faicuai]

19 minutes ago, Mazzspeed said:

A repository you don't contribute to.

This thread... is about... Maury's repo... and its related contributions.

 

This thread... is not about... and never was about... contributing to Bench64's repo. 

 

There is nothing, zero, nada, you have contributed to this thread, and this repo.

 

The above is irreducible and unquestionable. It is your factual record. 

 

 

 

 

 

 

[Mazzspeed]

21 minutes ago, Faicuai said:

This thread... is not about... and never was about... contributing to Bench64's repo. 

And yet you made it about Bench64's repo because you refused to accept certain benchmarking requirements.

 

Discussion over, let's get back on topic regarding a software repository and not the results of benchmarks from the software repository. Holy shit.

[Faicuai]

6 minutes ago, Mazzspeed said:

regarding a software repository and not the results of benchmarks

Maury... already has included... actual results... for many tests... on HIS repo.

 

No one else... except hin... decides what his repo... or its content... is all about.

 

Hoping that slow-and-easy males it easier to land back on HIS track.

 

 

[Faicuai]

@Maury Markowitz

 

Here are Altirra Extended Basic 0.15 result, booted from DOS XL (and Atarimax .car image) :

 

 

Math / FP library provided by AEB itself. 

 

It would be also great to see Actual-vs-Expected math. result added to the tests-summary, since the math test is designed to converge to a single value (golden-ratio approximation). This would be a useful enhancement over the original version. which hides the implied differences of precision... and precision is VERY important.

[Mazzspeed]

58 minutes ago, Faicuai said:

Maury... already has included... actual results... for many tests... on HIS repo.

Oh my God, the manipulation is amazing. We're talking about Bench64, Bench64. Therefore, Maury can post whatever he wants on his repo with everything bundled together (nothing against Maury, in fact I feel for the bloke on these forums). The fact remains that in the context of the rules as outlined by the author of the software on their own GitHub page (ie: scruss) your results are invalid.

 

Now give it a rest.

Edited June 12, 2021 by Mazzspeed

[Faicuai]

8 minutes ago, Mazzspeed said:

Oh my God, the manipulation is amazing. We're talking about Bench64, Bench64.

Already... included... there:

 

BASIC | System | Math lib | for | goto | gosub | if | fn | maths | string | array | overall |
		|--------------------|----------------------|----------|-----|------|-------|-----|-----|-------|--------|-------|---------|
		| Atari Rev C | Atari800XMac 6.0 | Atari | 31 | 21 | 24 | 49 | 46 | 33 | 53 | 56 | 35 |
		| Atari MS BASIC II | Atari800XMac 6.0 | MS BASIC | 94 | 47 | 101 | 96 | 92 | 142 | 102 | 90 | 88 | ***
		| BASIC XL 1.03 | Atari800XMac 6.0 | Atari | 35 | 27 | 28 | 59 | 62 | 33 | 74 | 80 | 42 |
		| BASIC XL 1.03 fast | Atari800XMac 6.0 | Atari | 100 | 140 | 129 | 152 | 105 | 33 | 113 | 118 | 91 |
		| BASIC XE 4.1 | Atari800XMac 6.0 | Atari | 37 | 24 | 27 | 65 | 60 | 33 | 60 | 75 | 40 |
		| BASIC XE 4.1 ext | Atari800XMac 6.0 | XE | 39 | 29 | 29 | 67 | 80 | 123 | 81 | 105 | 53 |
		| BASIC XE 4.1 fast | Atari800XMac 6.0 | XE | 130 | 176 | 145 | 206 | 167 | 128 | 129 | 177 | 153 |
		| Turbo-BASIC XL 1.5 | Atari800XMac 6.0 | Turbo | 182 | 154 | 155 | 256 | 154 | 137 | 151 | 169 | 164 |
		| Basic++ 1.08 | atari++ 1.83 | | 138 | 118 | 121 | 183 | 106 | 100 | 155 | 131 | 127 |
		Notes:
		*** For scruss runs, Atari MS BASIC was tested using the original C64 code, and is therefore not directly compariable to the Atari compatible BASICs. It is, however, suitable for comparison to other MS dialects found on other machines.

From this point on, we will not talk about anything else (whatsoever) that does not relate to expanding that results-table with ANY and ALL viable interpreters and FP packs that meet his own criteria.

 

More coming. We have not tested yet the fastest FP library, yet.

[Mazzspeed]

17 minutes ago, Faicuai said:

Already... included... there:

 

BASIC | System | Math lib | for | goto | gosub | if | fn | maths | string | array | overall |
		|--------------------|----------------------|----------|-----|------|-------|-----|-----|-------|--------|-------|---------|
		| Atari Rev C | Atari800XMac 6.0 | Atari | 31 | 21 | 24 | 49 | 46 | 33 | 53 | 56 | 35 |
		| Atari MS BASIC II | Atari800XMac 6.0 | MS BASIC | 94 | 47 | 101 | 96 | 92 | 142 | 102 | 90 | 88 | ***
		| BASIC XL 1.03 | Atari800XMac 6.0 | Atari | 35 | 27 | 28 | 59 | 62 | 33 | 74 | 80 | 42 |
		| BASIC XL 1.03 fast | Atari800XMac 6.0 | Atari | 100 | 140 | 129 | 152 | 105 | 33 | 113 | 118 | 91 |
		| BASIC XE 4.1 | Atari800XMac 6.0 | Atari | 37 | 24 | 27 | 65 | 60 | 33 | 60 | 75 | 40 |
		| BASIC XE 4.1 ext | Atari800XMac 6.0 | XE | 39 | 29 | 29 | 67 | 80 | 123 | 81 | 105 | 53 |
		| BASIC XE 4.1 fast | Atari800XMac 6.0 | XE | 130 | 176 | 145 | 206 | 167 | 128 | 129 | 177 | 153 |
		| Turbo-BASIC XL 1.5 | Atari800XMac 6.0 | Turbo | 182 | 154 | 155 | 256 | 154 | 137 | 151 | 169 | 164 |
		| Basic++ 1.08 | atari++ 1.83 | | 138 | 118 | 121 | 183 | 106 | 100 | 155 | 131 | 127 |
		Notes:
		*** For scruss runs, Atari MS BASIC was tested using the original C64 code, and is therefore not directly compariable to the Atari compatible BASICs. It is, however, suitable for comparison to other MS dialects found on other machines.

From this point on, we will not talk about anything else (whatsoever) that does not relate to expanding that results-table with ANY and ALL viable interpreters and FP packs that meet his own criteria.

 

More coming. We have not tested yet the fastest FP library, yet.

Well in that case, we also haven't tested a C128 running in C64 mode with screen blanked at 2Mhz and we haven't included the Ultimate 64 at 48Mhz as they both conform to MS BASIC. Furthermore, it's basically implied that only your MS BASIC run is comparable to other MS BASIC variants, your other runs are interesting but outside the context of the benchmark and are therefore fairly meaningless.

 

But you keep going on your little crusade wearing blinkers.

 

Furthermore, you're still referring to the wrong repo.

Edited June 12, 2021 by Mazzspeed

[Faicuai]

6 minutes ago, Mazzspeed said:

your other runs are interesting but outside the context of the benchmark and are therefore fairly meaningless.

What you think is irrelevant and immaterial, unless you would like to contribute with the Repo.

 

Try to sink that in, and also try to accept that I don't care the slightest bit for any other system that is not subject to Maury´s repo. scope.

 

The good news is that you can go ahead and either create your OWN repo. or enhance Bench64´s results with ANY of yours, in ANY WAY you want, which seems more important to you than supporting Maury´s  efforts.

 

 

[Mazzspeed]

26 minutes ago, Faicuai said:

What you think is irrelevant and immaterial, unless you would like to contribute with the Repo.

 

Try to sink that in, and also try to accept that I don't care the slightest bit for any other system that is not subject to Maury´s repo. scope.

 

The good news is that you can go ahead and either create your OWN repo. or enhance Bench64´s results with ANY of yours, in ANY WAY you want, which seems more important to you than supporting Maury´s  efforts.

 

 

Did you know that computers are inherently quite stupid machines? You have to break instructions down into more and more simplistic terms, until the code is so simple that even a computer can understand it....

 

Kind of reminds me of somethin...I mean, ah never mind, you won't understand. I'm done here, this is fucking retarded. I contributed to the silly benchmark using a PAL C64 that benched faster than an NTSC C64 with the screen blanked, meaning the results are invalid. What am I supposed to do, bench my A8 and create even more invalid results?

 

Edited June 12, 2021 by Mazzspeed

[Faicuai]

10 minutes ago, Mazzspeed said:

I'm done here, this is fucking retarded.

Don't forget to close the door, please.... 

 

Bon voyage!

[+Stephen]

16 hours ago, Faicuai said:

You can now go back to your kiddie computer

That right there sums up your attitude.  You think the C64 is a shit kiddie computer and the A8 is some saviour of the universe.  You do realize both machines are 40 year old "kiddie computers" at this point right?  You are stuck with the mentality of a 7 year old child on a playground and you look like a sad joke because of it.

[+Stephen]

15 hours ago, Faicuai said:

What really matters is what the creator of that Repo (Maury) truly wishes to accomplish with whatever gets loaded / shown there

And as we've stated multiple times, your DMA off results are invalid according to the author.  Yet you keep harping on.

[Faicuai]

1 minute ago, Stephen said:

That right there sums up your attitude.

iYou are free to construe it in any (pointless) way you may wish.

 

Enjoy!

[Faicuai]

1 minute ago, Stephen said:

And as we've stated multiple times

You are free to state them another million times, if you find that exhilarating. 

 

Have a blast!

[+Stephen]

2 minutes ago, Faicuai said:

You are free to state them another million times, if you find that exhilarating. 

 

Have a blast!

You obviously get a thrill from it Rain Man.  Back on ignore you go.

[Faicuai]

6 minutes ago, Stephen said:

Back on ignore you go.

Sweet !!!

 

What an honor!

 

??

[Maury Markowitz]

On 6/11/2021 at 7:16 PM, drpeter said:

That's definitely something to agree on.  Many would still take an 800 over any of the later models, which were predominantly exercises in reducing manufacturing costs.

Except the 1200XL, which may have indeed done that but then had a higher MSRP. What did Atari think they were selling users for all that extra money? 16k you could only use in certain cases?

 

I am absolutely convinced that if the 1200 had an 80-column mode, even monitor port only, it would have been the most popular machine in the lineup. Even failing that, one could immediately understand what you were getting for you money, even if it was even more money.

 

I still consider the 1200 to be the straw that broke the camel's back.

[+Stephen]

7 minutes ago, Maury Markowitz said:

Except the 1200XL, which may have indeed done that but then had a higher MSRP. What did Atari think they were selling users for all that extra money? 16k you could only use in certain cases?

 

I am absolutely convinced that if the 1200 had an 80-column mode, even monitor port only, it would have been the most popular machine in the lineup. Even failing that, one could immediately understand what you were getting for you money, even if it was even more money.

 

I still consider the 1200 to be the straw that broke the camel's back.

Well, it certainly would have been nice if sometime in the 15 years on the market the chipset saw at least one upgrade.  Conversely, I guess we can say that is was such a powerful chipset it ran strong for 15 years without any upgrades.

[+Larry]

So I just read this thread.  Well, I am interested in benchmarks, and as I understand the results (for Turbo Basic XL), I got an overall index of 164.  (I put in my TBXL cartridge, turned on the computer, and ran the BM.)  So this means that Turbo Basic on a A8 ran it 1.64X as fast as a typical C64 with its Basic.  Is that correct the correct interpretation?