RamseyConfig released!

[SpeedGeek]

RamseyConfig 1.5 ©SpeedGeek 2022

INTRODUCTION:
RamseyConfig allows you to view and change the config options of
the custom chip "Ramsey". This chip controls the Fast RAM in
A3000 & A4000 Amiga's. The other programs "SetRamsey" and
"Ramsey" have been around for sometime now, but with no bug
fixes, missing features or just supporting one version of Ramsey
these programs have serious limitations.

FEATURES:
- Small, fast and pretty reliable¹
- Shows Ramsey version² and config settings
- Config settings can be changed
- Dynamically manages Page mode setting for all users

REQUIREMENTS:
- Amiga 3000 or 4000
- OS2.0+ (maybe it works under OS1.3, but please no bug reports)

DISCLAIMER:
Use at your own risk. No warranty expressed or implied, etc.

NOTES:
¹Reliable IMHO means the program should not crash or mess up
the Ramsey config register because of bugs which can be
prevented (e.g. Page mode is invalid on version $D even for A3000
users). But it's certainly OK for the user to enter a valid config
option which may not be supported on a certain system (see WARNINGS) thereby, possibly causing any or all of the above to happen.
RamseyConfig's argument list has the safest options at the
beginning of the list and the more risky options at the end of
the list so nobody should use them by accident!

²RamseyConfig simply reports the version register verbatim so if
this doesn't match the expected, official or chip printed
version number please don't send me version related bug reports!

UPDATE INFO:
The new ReadArgs() parsing code (since v1.2) now accepts up to
4 arguments but you must use the full argument name. Also,
delimiters must be spaces.

HISTORY:
v1.0 First release
v1.1 Some useful improvements
- Added code to parse multiple arguments
- Simplified argument message code
v1.2 RamseyConfig now works more like a "Standard" Shell tool!
- Replaced custom argument parsing code with ReadArgs()
- Replaced paired SuperState()/UserState() calls with single
Supervisor() calls
v1.3 Added code to disable Burst and Page mode for A4000 users since they can't use either of them.
- New feature, added (version dependent) code to show Ram width or Skip mode.
v1.4 A4000 Bug Fix
- Replaced SDMAC test with A3000 bonus code test.
v1.5 Replaced A3000 bonus code test with Lisa test. In most cases,
hardware model detection should be more reliable than
software model detection.

[hooverphonique]

nice one.. what about the bit for 4/5 cycle selection?

[TheBilgeRat]

Originally Posted by SpeedGeek

RamseyConfig 1.0 ©SpeedGeek 2015

INTRODUCTION:
RamseyConfig allows you to view and change the config options of
the custom chip "Ramsey". This chip controls the Fast RAM in
A3000 & A4000 Amiga's. The other programs "SetRamsey" and
"Ramsey" have been around for sometime now, but with no bug
fixes, missing features or just supporting one version of Ramsey
these programs have serious limitations.

FEATURES:
- Small, fast and pretty reliable¹
- Shows Ramsey version² and config settings
- Config settings can be changed
- Dynamically manages Page mode setting for both versions

REQUIREMENTS:
- Amiga 3000 or 4000
- OS2.0+ (maybe it works under OS1.3, but please no bug reports)

NOTES:
¹Reliable IMHO means the program should not crash or mess up
the Ramsey config register because of bugs which can be
prevented (e.g. Page mode is invalid on version $D). But it's
certainly OK for the user to enter a valid config option which
may not be supported on a certain system (see WARNINGS) thereby,
possibly causing any or all of the above to happen.
RamseyConfig's argument list has the safest options at the
beginning of the list and the more risky options at the end of
the list so nobody should use them by accident!

²RamseyConfig simply reports the version register verbatim so if
this doesn't match the expected, official or chip printed
version number please don't send me version related bug reports!

HISTORY:
v1.0 first release

Neat! So, why would someone want to twiddle Ramsey?

[SpeedGeek]

Originally Posted by hooverphonique

nice one.. what about the bit for 4/5 cycle selection?

I assume your asking about the skip bit on version $F(G?) or -07 Ramsey. IIRC this is the bit set by "SpeedRamsey" and similar programs. The reason it's not supported in this release is simply because it's pretty "Unreliable" and would conflict with one of the main features of RamseyConfig.

There are just to many cases where it can fail... You need 60ns DRAM installed, but even you have 60ns DRAM it still may not work with all banks fully populated... It won't work if the mobo is overclocked... It may be unstable or not work if the +5 Volt supply is low or the RAM sockets are in poor condition, etc.

If there are many requests, I may reconsider it for the next release but it's a profoundly different situation than supporting Page or Burst mode which do work reliably with the correct type of RAM installed.

Originally Posted by TheBilgeRat

Neat! So, why would someone want to twiddle Ramsey?

Apparently, you didn't RTFM the docs of RamseyConfig or any of it's predecessor programs (or search on the many Amiga forums where this topic has been discussed in greater detail).

But the simple answer is, you might be able to improve your Fast RAM performance if you twiddle Ramsey (just understand there may be some risk in doing that).

[TheDarkCoder]

Hi SpeedGeek, it's awesome to have a new and up-to date program to play with Ramsey. I am not an expert at all. However, as fortunate coincidence, in these days I am playing with the skip mode (4/5 cycles) in my 4000D.

Currently I am able to use it with 3 populated slot. Unfortunately I don't have a a fourth 60ns SIMM.

I use SpeedRamsey and Memutil to see if the RAM access is reliable.
I would love to see this setting supported also by your program, and it would be nice to integrate in your program a ram test tool.
Maybe you could implement in your program a "reliable mode" and a "dangerous " one which enable the setting.
Whatever you decide to do, well done and I will test your program ASAP

[SpeedGeek]

Originally Posted by TheDarkCoder

Hi SpeedGeek, it's awesome to have a new and up-to date program to play with Ramsey. I am not an expert at all. However, as fortunate coincidence, in these days I am playing with the skip mode (4/5 cycles) in my 4000D.

Currently I am able to use it with 3 populated slot. Unfortunately I don't have a a fourth 60ns SIMM.

I use SpeedRamsey and Memutil to see if the RAM access is reliable.
I would love to see this setting supported also by your program, and it would be nice to integrate in your program a ram test tool.
Maybe you could implement in your program a "reliable mode" and a "dangerous " one which enable the setting.
Whatever you decide to do, well done and I will test your program ASAP

Originally Posted by boing4000

If the Ramsey skip mode conflict with other feature, do not implement it.
For some boards it works fine (my own 4000D with 4x 60nS Simm) but others freez instandly or become instable after some warming time.

There are mentioned tools to set this bit (also GUI based one such as SetCPU). Better let user try this if really wanted. Its only one more line in the startup-sequence

I'll wait to see if more comments are made on this topic, but it looks like a separate "Dangerous" version of the program is the only solution which could satisfy everyone.

I don't think adding a RAM test is really practical for this type of program. The only really useful RAM test is to determine if Page or Static column RAM is installed (and that's exactly what the Bonus code in your Kickstart ROMs is supposed to do).

The best time to do this test is in early startup before any Fast RAM is added to the system or in use by other programs. The are 4 possible banks of Fast RAM and they should all be tested before either Page or Burst mode is enabled. That's problematic at best in a post boot scenario (expecting some RAM to be available in all 4 banks and then to disable all interrupts and other tasks while the test is being performed). Another problem is Ramsey version $D doesn't support Page mode so the test can't be done anyway.

As far as testing if the skip mode is working reliably, there are quite a number of programs available to do a standard RAM test already.

[SpeedGeek]

** NEWS UPDATE **

v1.1 Released!

Some useful improvements
- Added code to parse multiple arguments
- Simplified argument message code

UPDATE INFO:
The new multiple argument parsing code (since v1.1) has the
following limitations:
- Delimiters must be either spaces or commas
- The maximum number of arguments accepted is 4
- The first invalid argument terminates the argument search

[SpeedGeek]

** 2ND NEWS UPDATE **

v1.2 Released!
RamseyConfig now works more like a "Standard" Shell tool!
- Replaced custom argument parsing code with ReadArgs()
- Replaced paired SuperState()/UserState() calls with single
Supervisor() calls

UPDATE INFO:
The new ReadArgs() parsing code (since v1.2) now accepts up to
4 arguments but you must use the full argument name. Also,
delimiters must be spaces.

[SpeedGeek]

Originally Posted by amigoun

This is a message from amigoun at English Amiga Board ( http://eab.abime.net/index.php ). The English Amiga Board owners cannot accept any responsibility for the contents of the email.:

Hi SpeedGeek,

I have recently come to your RamseyConfig program and I decided to tweak the Ramsey in my A3000.
Some time ago I replaced the page-mode ZIPs in my A3000 with static-colums ZIPs that I assume is better speedwise. At least Sysinfo now shows 1.0 performance instead 0.9 before.
I think I could try some other tweaking, does the Ramseyconfig helps to speedup it more? What option to try you would recommend?
Also I would like to ask - the previous page-mode ZIPs were 60ns whereas the Static column is 70ns. What do you consider better and faster?

kind regards

Hi amigoun,

It appears you have already obtained the most significant performance improvement for the A3000 with Burst enabled (Static Column) RAM. This performance improvement is unfortunately NOT supported by nearly all 040/060 accelerator cards so there is some question if the cost can be justified for 68030 use only.

You can usually increase the refresh mode to the next higher refresh mode for a small performance tweak. There is no practical performance difference between 60-70 ns RAM unless you are willing to "Risk" skip mode as previously explained on this thread.

Best regards,
Kevin

[amigoun]

Great, thanks for the detailed explanation.

[SpeedGeek]

Okay, RamseyConfig is due for an update (Thanks to macce2 for helping me realize this).

PAGE mode will absolutely crash the machine for A4000 users. Therefore, this option should be disabled permanently in this case. However, BURST mode will not always crash the machine for A4000 users (even though they won't be able to use it).

This is a situation which IMO, requires user feedback. So, I will wait a while to see what kind of feedback is given before releasing the next update.

[macce2]

Very nice to hear that !

[SpeedGeek]

** 3RD NEWS UPDATE **

RamseyConfig 1.3 released!

Added code to disable Burst and Page mode for A4000 users since they can't use either of them.
- New feature, added (version dependent) code to show Ram width or Skip mode.

[macce2]

The new version of RamseyConfig immediately crashes my machine !

[SpeedGeek]

** 4TH NEWS UPDATE **

RamseyConfig 1.4 released!

v1.4 A4000 Bug Fix
- Replaced SDMAC test with A3000 bonus code test. Thanks again to macce2 for the bug report!

[macce2]

The new version seems to work.
Should the higher refresh mode necessarily improve general system speed ?!
If so, on what kind of operations should one especially notice the (marginal !?) improvement ..?! (If it is even noticeable )

[SpeedGeek]

** 5TH NEWS UPDATE **

RamseyConfig 1.5 released!

v1.5 Replaced A3000 bonus code test with Lisa test. In most cases,
hardware model detection should be more reliable than
software model detection.

@macce2

If you check the AIBB memory latency result you might see a decrease by .1. As previously posted, the refresh rate increase is a small tweak.

[macce2]

It seems running RamseyConfig with SKIP mode, WRAP and/or higher refresh mode
makes my system to crash after a while.
So, maybe this ain't bug report, per se.
I assume the default value of refresh mode to be R248 !?

[thebajaguy]

Just reporting in the system config, no issues, using v1.5.

Base System: A4000D with 16MB 70ns DRAM SIMMs, RAMSEY-07

GVP T-Rex I 68040/40MHz, 8MB

OS 3.2 ROM, OS 3.2.1 update after SetPatch (no library), reports:

RamseyConfig 1.5 ?SpeedGeek 2022
Ramsey version = $F
Page mode = 0
Burst mode = 0
Wrap mode = 0
Ram width = 0
Refresh rate = 238

Putting the v44 68040.library in, refresh rate adjusts to 380 after SetPatch is called.

Only other option to modify is Wrap. BusTest speeds do not vary against the Ramsey memory.


Also tested on an A3000/25 with 16MB 60ns SCRAM, Ramsey-04 ($D)/DMAC-02, with a BFG-9060/50MHz.

Refresh was set optimally to 380 after CPU library load. RAM width was reporting 1. Adjusting Burst or Wrap had no measurable effect in speed or stability.

[matt3k]

Just updated from 1.1 to 1.5 and having issues with it.

I have a 3000D with DMAC04 and Ramsey07.
With 1.1 I could set PAGE BURST R380 and all worked perfectly.

With 1.5 the system locks most times on a boot, I have to turn it off to get it to boot.

It always ignores the R380 and doesn't set it on boot. I can turn it on if I just run the program from a shell specifying R380.

The skip mode can't be set and won't turn on.

Happy to provide any other info to fix it.