Jump to content
IGNORED

Atari Disk Speed vs C64 (stock and modded)


bbking67

Recommended Posts

5 hours ago, Faicuai said:

And here's the absolute maximum we will be able to get out of SIO-attached .ATR images, still without PC-Link packet transmission, but provided here only as  a reference, since PBI-hSIO Bios is required (Divisor-0), plus support pf 512 Bytes/sec. on ATR format (possible with RespeQt PC server, for instance):

 

495E9DCE-3262-410C-9C17-0C2E413489AD.thumb.jpeg.21043956a2d42f527202a57007f61b7e.jpeg

 

 

I find divisor 0 via RespeQt server flaky at times. I know it's not the lead as I made it and put a lot of effort into it using the FT232 chipset with PROCEED connected.

 

As a result I tend to use divisor 5 most of the time as I find it to be a good compromise between speed and reliability. I downloaded the unfinished Prince Of Persia demo the other day, I had to drop speeds right down to stock to load xbios.com reliably.

 

There's also a number of points to mention regarding your benchmarking batch files:

 

1 - You disable DMA, which blanks the screen, something early Commodore fastloaders used to do and something many wouldn't find desirable in daily use without some form of hardware 80 column device. Doing so regarding the PBI HDD bench results in an improvement of around 20,000k/bits per second!

 

2 - PAL users won't achieve the speeds you show regarding the PBI HDD bench due to variances in CPU clock speed.

Edited by Mazzspeed
Link to comment
Share on other sites

17 minutes ago, Mazzspeed said:

I find divisor 0 via RespeQt server flaky at times.

Yes, it could be at times, especially without any signal-shaping / handling, such as Lotharek's SIO-hub (with which you will attain red-hot divisors 1/0 all day long).

 

Also, running your console session (E:) with DMA disabled (like XEP80, Bit3, etc.) will also support stability at higher speeds (lower divisors).

 

Divisors 5/4 with no other involved HW should be sustainable, all day long. XBIOS may or may not work with HSIO, though...

Edited by Faicuai
Link to comment
Share on other sites

16 minutes ago, Mazzspeed said:

PAL users won't achieve the speeds you show highlight regarding the PBI HDD bench due to variances in CPU clock speed.

I would expect PAL users to attain better speeds since the CPU in an NTSC system spends more time per second running the vertical blank interrupt handler.

Link to comment
Share on other sites

2 minutes ago, Faicuai said:

Yes, it could be at times, especially without any signal-shaping / handling, such as Lotharek's SIO-hub (with which you will attain red-hot divisors 1/0 all day long).

I'm only running one device on my SIO port directly to the USB port of my server machine, it shouldn't really make that much of a difference in this scenario?

Link to comment
Share on other sites

4 minutes ago, Faicuai said:

Yes, it could be at times, especially without any signal-shaping / handling, such as Lotharek's SIO-hub (with which you will attain red-hot divisors 1/0 all day long).

I must have gotten lucky here. I run divisor 0 100 per cent of the time for critical operations such as firmware updates on a variety of machines with no errors.

Link to comment
Share on other sites

5 minutes ago, flashjazzcat said:

I would expect PAL users to attain better speeds since the CPU in an NTSC system spends more time per second running the vertical blank interrupt handler.

This is the fastest speed I can achieve regarding RWTEST using the exact same batch file used by Faicuai, in comparison he gets nearly 90,000B/sec using SIDE2.

 

Unless there's an optimization difference between SIDE2/3, the only thing I can put it down to is clock speed variances?

 

fC4c2He.jpg

Link to comment
Share on other sites

7 minutes ago, flashjazzcat said:

I must have gotten lucky here. I run divisor 0 100 per cent of the time for critical operations such as firmware updates on a variety of machines with no errors.

My system isn't 'that' bad re: Divisor 0, most of the time it's fine. However I wouldn't trust it at that speed regarding a firmware update.

Link to comment
Share on other sites

1 hour ago, Mazzspeed said:

This is the fastest speed I can achieve regarding RWTEST using the exact same batch file used by Faicuai, in comparison he gets nearly 90,000B/sec using SIDE2.

 

Unless there's an optimization difference between SIDE2/3, the only thing I can put it down to is clock speed variances?

There's a fairly obvious difference between SIDE2 and SIDE3: the former uses CF cards and the latter SD cards. The drivers are completely different and though the speed of both is primarily governed by the CPU, the SD card protocol bears little resemblance to ATA and SD cards themselves can vary quite significantly in terms of what is effectively seek latency (i.e. the delay between issuing a command and receiving acknowledgement and data).

 

Even two identical setups (SIDE2 or 3) may return different results depending on the media being used.

1 hour ago, Mazzspeed said:

My system isn't 'that' bad re: Divisor 0, most of the time it's fine. However I wouldn't trust it at that speed regarding a firmware update.

Sure: if I didn't get stable 127Kb/s in every machine which comes through my hands, I wouldn't be running divisor 0 on them, and as far as I'm concerned, it either works perfectly or it doesn't work at all. There are many variables here too: the host PC, cable length, SIO caps (which often need to be removed), FTDI chip, etc. I use a Lotharek SIO2PC, but prior to this a self-built device which performed identically well. The one I built before that was flaky as Hell, presumably because the FTDI chip was counterfeit (I guess).

 

I would waste an awful lot of time waiting for 512KB firmware files to transfer over SIO if divisor 0 wasn't solidly reliable.

Edited by flashjazzcat
Link to comment
Share on other sites

1 hour ago, flashjazzcat said:

Sure: if I didn't get stable 127Kb/s in every machine which comes through my hands, I wouldn't be running divisor 0 on them, and as far as I'm concerned, it either works perfectly or it doesn't work at all. There are many variables here too: the host PC, cable length, SIO caps (which often need to be removed), FTDI chip, etc. I use a Lotharek SIO2PC, but prior to this a self-built device which performed identically well. The one I built before that was flaky as Hell, presumably because the FTDI chip was counterfeit (I guess).

Does the 600XL have the caps that need to be removed?

 

6 minutes ago, CharlieChaplin said:

Maybe my eyes are fooling me, but I think I only see 9075,xyz B/sec. on Faicuai's screenshot above (post #50), not 90,000 B/sec. or am I wrong ?!?

 

Wrong benchmark, look at the previous page for the PBI HDD bench.

 

1 hour ago, flashjazzcat said:

There's a fairly obvious difference between SIDE2 and SIDE3: the former uses CF cards and the latter SD cards. The drivers are completely different and though the speed of both is primarily governed by the CPU, the SD card protocol bears little resemblance to ATA and SD cards themselves can vary quite significantly in terms of what is effectively seek latency (i.e. the delay between issuing a command and receiving acknowledgement and data).

 

Even two identical setups (SIDE2 or 3) may return different results depending on the media being used.

Valid point. I doubt the difference would be in any way noticeable in real world usage, I just found it interesting. In fact, I assume that due to it's DMA implementation, in normal real world usage the SIDE3 should have a slight edge shouldn't it?

Link to comment
Share on other sites

13 minutes ago, CharlieChaplin said:

So, R/W-Test on the A8 gives:

 

- max. 9000 B/sec. for a SIO device

- max. 90,000 B/sec. for a HDD device

 

Or are there higher speeds (for SIO or HDD) available?

 

With the screen blanked. With normal ANTIC enabled usage you can chop ~20,000 B/sec off the PBI HDD speed using normal formatting.

  • Like 1
Link to comment
Share on other sites

1 hour ago, flashjazzcat said:

which often need to be removed

But the funny thing is that there is not (any) removal of (anything) happening, here.

 

Aiming at stable, Divisor 1/0, all-day-long, 7x24, on a COMPLETELY STOCK machine is the name of the game!

 

All #s posted above come from stock (SIO host interface) 1980 CTIA and 1983 GTIA 800's. Surgery-free babies... 8-)

Link to comment
Share on other sites

I'm just benchmarking an SIO attached disk at divisor 0 using high speed SIO BIOS using an ATR at 512 bytes/sec and I get nowhere near the speeds Facuai is getting via RespeQt.

 

zc5aQt8.jpg

 

I don't know what I'm doing wrong here:

 

Pil6bah.jpg

 

vt57dfv.jpg

 

5Dv2fm6.jpg

 

 

Edited by Mazzspeed
Link to comment
Share on other sites

1 hour ago, CharlieChaplin said:

So, R/W-Test on the A8 gives:

 

- max. 9000 B/sec. for a real (?) SIO device

- max. ??? B/sec. for an emulated SIO2xyz device

- max. 90,000 B/sec. for a HDD device

 

Or are there higher speeds (for real/emulated SIO or HDD) available?

 

  • 4,500 Bytes/sec from Indus/GT Track&Sector buffer (RamCharger), NO magnetic-surface read, 256 Bytes/sec in double-density buffered sectors (forget about the. 69Kbps and all that line-rate baloney). 8,000 bytes/sec will be achieved INTERNALLY on CP/M (magnetic-surface to RAM Charger 64KB ram).
  • 10,000 Bytes/sec. over SIO (but requiring SIO device drive supporting 512 Bytes/sec, and flawless interfacing and response at Divisor 0, NO PC-Link yet)
  • 90,000 Bytes/sec on SanDisk Extreme III 4GB CF card, and running system in console mode from SDX (if interactive session required) via XEP80 or Bit3 (on NTSC). If not, that speed can be achieved via OS control of DMA during any inert I/O operation.
  • 105,000, Bytes/sec on same card, same conditions, but direct sector reads (free of SDX overhead, about 14%)

 

That is about it...

Edited by Faicuai
Link to comment
Share on other sites

7 minutes ago, Mazzspeed said:

I'm just benchmarking an SIO attached disk at divisor 0 using high speed SIO BIOS using an ATR at 512 bytes/sec and I get nowhere near the speeds Facuai is getting via RespeQt.

 

zc5aQt8.jpg

 

I don't know what I'm doing wrong here:

 

vt57dfv.jpg

 

5Dv2fm6.jpg

 

 

My SIO chain:

 

A8 Host => SIO Hub 1x4 (Lotharek) => SIO-to-USB (Lotharek)plugged in SIO hub port 4 => USB port on PC (RespeQt in Windows-10/Pro)

 

Divisors 1/0 all-day-long, even from 1980 CTIA 800. That also includes SDrive NUXX on port #2 of SIO-hub, and the last high speed firmware update.

Edited by Faicuai
Link to comment
Share on other sites

1 minute ago, Faicuai said:

My SIO chain:

 

A8 Host => SIO Hub 1x4 (Lotharek) => SIO-to-USB (Lotharek)plugged in SIO hub port 4 => USB port on PC (RespeQt in Windows-10/Pro)

 

Divisors 1/0 all-day-long, even from 1980 CTIA 800.

But is there anything wrong with my settings there? As RespeQt isn't reporting any transmission errors and I get nowhere near 10,000 B/sec at divisor 0 over SIO.

Link to comment
Share on other sites

3 minutes ago, Mazzspeed said:

But is there anything wrong with my settings there? As RespeQt isn't reporting any transmission errors and I get nowhere near 10,000 B/sec at divisor 0 over SIO.

It shows effective speed of 19,200 on bottom-rght corner.

 

Please, run "reset" command on provided 16MB .atr or simply invoke BIOS and get back to SDX with ESC. Make sure the drive logged in RespeQt is also enabled as high-speed on BIOS.

 

Try again only then.

 

Edited by Faicuai
Link to comment
Share on other sites

Just now, Faicuai said:

It shows effective speed of 19,200 on bottom-rght corner.

 

Please, run "reset" command on provided 16MB .atr or simply invoke BIOS and get back to SDX with ESC. Make sure the drive logged in RespeQt is also enabled as high-speed on BIOS.

No matter what I do, that speed doesn't change - How do I change it?

 

HSIO BIOS settings below:

 

Pil6bah.jpg

Link to comment
Share on other sites

5 minutes ago, Mazzspeed said:

No matter what I do, that speed doesn't change - How do I change it?

  1. Does your config.sys shows DEVICE SIO or DEVICE SIO/A
  2. If you can't change the speed. you have a HW interfacing problem on your SIO chain, unfortunately (one active element is not negotiating or not setting Divisor-0 as speed-rate, thus defaulting back to 19.2Kbps, thus no errors...)

 

Try manually disconnecting and reconnecting the link on RespeQt and issuing a DIR command from SDX prompt.

Edited by Faicuai
Link to comment
Share on other sites

4 minutes ago, Mazzspeed said:

Config.sys shows DEVICE SIO. If I try to add DEVICE SIO/A I get config errors when loading drivers on boot.

 

Does the 600XL have the capacitors on the SIO port that need to be removed? As I couldn't see them when I was working on the device.

DO NOT use SIO/A. Fine if just SIO. Just checking you were truly running I/O through PBI BIOS.

 

You should not need to remove anything from your 600XL to achieve those speeds, provided you build the right SIO chain.

 

My inner psychopath violently refuses to perform unnecessary HW changes on any of my fleet's units, including my 800XLs / U1MB / SIDE2, btw.?

 

 

Edited by Faicuai
Link to comment
Share on other sites

1 minute ago, Faicuai said:

DO NOT use SIO/A. Fine if just SIO. Just checking you were truly running I/O through PBI BIOS.

 

You should not need to remove anything from your 600XL to achieve those speeds, provided you build the right SIO chain.

 

 

I rectified the config.sys settings...

 

The SIO chain is perfect, RespeQt is simply not allowing those speeds. The SIO cable even uses a shielded ground.

 

fYh6EGh.jpg

Link to comment
Share on other sites

1 minute ago, Mazzspeed said:

I rectified the config.sys settings...

 

The SIO chain is perfect, RespeQt is simply not allowing those speeds. The SIO cable even uses a shielded ground.

 

fYh6EGh.jpg

 

Manually throttle back divisor by divisor, until you find the first sustained hand-shake.

 

There is a bottleneck, somewhere in there.

Edited by Faicuai
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...