Any Success Stories with Rapidus?

[drac030]

"Why not"? Perhaps because you are the only one who apparently is experiencing serious problems.

 

In my eyes, you have lost the credibility once you refused to send any of your computers with Rapidus to anyone else for verification (Pasiu, Lotharek). For me this simply means that you do not want the problem to be fixed.

Edited June 16, 2016 by drac030

[flashjazzcat]

For me this simply means that you do not want the problem to be fixed.

Seriously? You think I'd rather spend weeks pottering around trying to diagnose these issues than simply sit down and enjoy the device?

 

If my statements carry no weight because I'm reluctant to entrust machines to international couriers (and no-one aside from yourself suggested this yet), then fair enough. I was asked to fix some issues once I had the actual hardware in my hands, and I've had no success with that, and a number of other problems I haven't been able to overcome. The whole point of the exercise was to get things fixed, but if there's nothing broken, I'm wasting my time. Hopefully the U1MB issues which can be replicated elsewhere will be fixed by the respective hardware people some time. I'll check in again in a few months, perhaps.

[danwinslow]

Drac, Jon is not trying to trash Rapidus's reputation. He has done a lot of posting about stability problems, but he was trying to get them fixed, not just say 'Rapidus sucks'. It does raise awareness of problems though, to have such extended discussions, so I get where you're coming from. But Jon's credibility is unquestioned by me and I think most people.

 

I am glad to see this thread, because often people only talk about problems. That goes for many things, not just obscure Atari hardware. The ones that DON'T have any problems don't usually post 'I have had no problems!' and so there we are. I am getting a couple of Rapidii installed shortly and I'll be glad to report on having success. But, like Jon, if I have problems I'll surely post that too.

[flashjazzcat]

Thanks Dan. I have nothing but unreserved respect and admiration for Konrad's work and I know from experience how frustrating it can be as a software developer when one or two users are having persistent problems. The temptation may be to suggest that the user can't solder three wires correctly or is "doing it wrong", but while I'm not an expert in electronics, I'm a competent hardware installer and have been troubleshooting these kinds of problems with the remote help of experts like Hias and Candle for a number of years. But never mind.

[drac030]

In my eyes, you have lost the credibility once you refused to send any of your computers with Rapidus to anyone else for verification (Pasiu, Lotharek). For me this simply means that you do not want the problem to be fixed.

Is my English not understandable?

[Marius]

Please calm down.

 

Rapidus is in development phase. The early adopters have their Rapidus installed, and -of course- in the field the first problems arise. That is 100% normal.

 

Unfortunately the userbase of Rapidus is way too small to have any conclusions about the product yet.

 

I don't have Rapidus, so I can not write about it. But... I know myself, and one thing that is obvious... as soon as I start using my atari 8bit I almost ALWAYS run into bugs that were not seen before. It's a personal curse I believe.

 

I have a pretty unconventional use of my a8 stuff, so it is very likely that I run into bugs and trouble, that a developer did overlook because he couldn't even imagine that someone wants what I do. I think that my reports are interesting and useful, just because of that.

 

Now back to this issue...

 

Isn't there a slight possibility that FJC is in some way like me... that he found an issue, that perhaps other people did not run into SO FAR because Rapidus is new, and not all people are using their atari's 24/7 (I know FJC does) ... and people perhaps mostly played with Rapidus in NON-classic mode?

 

I think everybody here wants Rapidus to be as good as possible, and I think FJC's experience is very valuable. Sending his stuff to the hardware team would be -in this phase- not a very wise thing todo, since it is way smarter to keep the failing setup at the user where the problems occur. The chances are VERY high that the hardware team will not have the issue, since the hardware team uses the Rapidus not in the way FJC does.

 

But first of all: please guys... keep it cozy

[drac030]

Isn't there a slight possibility that FJC is in some way like me... that he found an issue, that perhaps other people did not run into SO FAR because Rapidus is new, and not all people are using their atari's 24/7 (I know FJC does) ... and people perhaps mostly played with Rapidus in NON-classic mode?

There is not slight, but AMPLE possibility that the things are like you describe. The device is new and I am sure I would behave exactly like him, when I got a board advertised like working while it would not work for me. That is all okay.

 

It stops to be okay when a) I installed the board myself in my computer, b) AND it does not work for me no matter what, c) AND it does for other people, d) AND I refuse to allow anyone to verify my installation.

 

That last part of the equation, when true, is crucial.

[Marius]

There is not slight, but AMPLE possibility that the things are like you describe. The device is new and I am sure I would behave exactly like him, when I got a board advertised like working while it would not work for me. That is all okay.

 

It stops to be okay when a) I installed the board myself in my computer, b) AND it does not work for me no matter what, c) AND it does for other people, d) AND I refuse to allow anyone to verify my installation.

 

That last part of the equation, when true, is crucial.

 

a) If it is not allowed to install the board yourself, it should not be sold in that way. Many people install this kind of stuff without any help. The installation procedure is straight forward, and with FJC history of installation experience this statement is (almost) an insult. The website of lotharek even claims that it is a SIMPLE installation.

 

b) nobody says it does not work no matter what, fjc found some odd behavior in different computers. Odd behavior that is seen more in upgrades today.

 

c) the userbase and the time people were able so far to experience with Rapidus is way too small to conclude that it works for other people. I am not bashing this product, but from what I have read so far it seems more trouble than success. Which still makes sense for me, since it is a complete new product, which can never be beta tested very well with such a small userbase.

 

d) he does not refuse anyone to investigate, but there are some (in my opinion valid) reasons for him not to ship it out (at this very moment)

 

I think that FJC is a very valuable source of information in this problem, and I find it a bit harsh how you react on his findings. You come with conclusions that he does not WANT this problem to be fixed. It seriously shocks me that you keep repeating that.

 

It is definitely not a matter of NOT WANTING.

Edited June 17, 2016 by ProWizard

[flashjazzcat]

There's no need to verify the installation. I'm not a bloody idiot, and I'm more than a bit peeved at the suggestion I lack the competence to install the device in a way which does not require verification. At no point have Pasiu or Lotharek (during many email exchanges on the subject) called my ability to install the device into question, and as the hardware designer and vendor respectively, theirs is the opinion which matters. Similarly, I'm able to advise on Ultimate 1MB installations, not because I wrote a BIOS for the device, but because I have installed about twenty of them over the years in machines people sent me to upgrade. And when a second opinion is required (from, for example, Hias), somehow solutions have been found via email or private message (or via this forum). This is the age of the Internet after all, and collaborative troubleshooting is possible without everyone sitting in the same room.

 

Anwyay: before I go, since the latest 130XE tests showed issues clearly focused on 6502 mode, I did what Prowizard had talked about a while ago and just now trained a cold stream of air (from my wife's hair dryer) directly on the 6502C once the reboot issue has started up. Reboot issue immediately disappeared once 6502 was cooler. Left it again for five minutes. Reboot issue returns.

 

So, assuming I'm competent enough to have pointed the hair dryer at the correct component and my observations are not clouded by the symptoms of mental illness: looks like the 6502 overheats!

 

Since I observed something similar on four machines with eight different CPUs, it might be useful if anyone else was able to observe this issue if they have the time or inclination to run the device in Classic mode for more than five minutes. If everything works, then wonderful. Either way, the information might prove useful.

 

Overheating not mean that Rapidus cannot be used (I'm off to look for some low-profile heatsinks which will fit in the narrow space between the top of the CPU and the 130XE's lid), nor that it isn't a great device. It would simply mean that the 6502 may overheat. If eight CPUs overheat on two different Rapidus boards in four different Ataris in one particular house, there's maybe a small chance the issue isn't unique.

 

I understand that Beetle already has a cooling fan aimed at his board, so perhaps the idea of additional cooling isn't as crazy as it sounds.

 

Anyway - I hope the findings are of some use, and I'm sorry that a simple canvassing of Rapidus owners for much needed information proved such an inconvenience.

Edited June 17, 2016 by flashjazzcat

[pirx]

hey! bodge up a heat pipe from this 6502 to gita and antic and enjoy the DGF effect!

win-win-win situation that is!!!

 

http://atarionline.pl/v01/index.php?subaction=showfull&id=1361631036&ucat=1&ct=nowinki

Edited June 17, 2016 by pirx

[pirx]

Whoa!

http://us4.campaign-archive1.com/?u=7b2a36290b1ce8b46c9316bc2&id=c057c55984

[danwinslow]

https://www.sparkfun.com/products/10080

https://www.sparkfun.com/products/11510

https://www.amazon.com/dp/B00GWFMRDU/ref=pd_lpo_sbs_dp_ss_3?pf_rd_p=1944687462&pf_rd_s=lpo-top-stripe-1&pf_rd_t=201&pf_rd_i=B007XACV8O&pf_rd_m=ATVPDKIKX0DER&pf_rd_r=JQRP8X03NRHR9S7MA0XP

http://www.kr4.us/small-heatsink.html?gclid=CjwKEAjw7qi7BRCvsr3N58GvsTkSJAA3UzLvNkKS531_ZT-aIxztkSyOqtX0UBx3p8omzSvj_4eZWhoCZiPw_wcB

 

Etc.

[flashjazzcat]

Current working solution:

 

 

Heatsink is a passive chipset cooler taken from a dead PC motherboard. Front corner had to be filed down slightly in order to get the 130XE's lid down, but it works reliably and providing stock OS is used, Ultimate 1MB operation appears trouble-free, even at 20MHz. Reboot on reset issue is completely banished. Address line pull-ups are removed and only remaining modification is extra GND wire from U1MB to the Atari motherboard, although who knows if this is still necessary. It does no harm.

 

I don't have temperature measuring equipment here, but the cooler experiences considerable heat build up when Rapidus core is present (i.e. when 65C816 is the primary CPU and RDY is negated on the 6502C). Warmth can be felt on the plastic casing directly above the processor, although not sufficient cause melting. Thermal effects on the 6502C itself are unknown.

[HiassofT]

Instead of doctoring around the symptoms it would be good to find out why exactly the 6502 is overheating and fix that.

 

For example it is well known that the original NMOS 6502 can overheat if reset is asserted too long

http://forum.6502.org/viewtopic.php?t=1927

 

Not sure if the same applies to the Atari CPU and even fewer ideas what Rapidus might be doing.

 

so long,

 

Hias

Edited June 22, 2016 by HiassofT

[flashjazzcat]

Instead of doctoring around the symptoms it would be good to find out why exactly the 6502 is overheating and fix that.

 

For example it is well known that the original NMOS 6502 can overheat if reset is asserted too long

http://forum.6502.org/viewtopic.php?t=1927

Agreed, and thanks for the link Hias. I mentioned to Lotharek "some stuff" I had read about 6502 overheating in these circumstances, but could not locate it again.

[flashjazzcat]

Heatsink fell off (superglue failed), but these arrived yesterday.

 

[+Larry]

Jon, that looks nice and hopefully will completely solve your issues. Ever thought of setting up the only liquid cooled Atari?

 

-Larry

[flashjazzcat]

Jon, that looks nice and hopefully will completely solve your issues. Ever thought of setting up the only liquid cooled Atari?

Without a doubt (since the only remaining issue is that of durable adhesion). Dan already suggested a Peltier in post 37, but I had not yet considered a water-cooled rig.

 

I must add that I was recently informed (by Pasiu) that the 6502C provides the Phi2 signal at all times, even when the 65C816 is the primary CPU. The 6502C is then held in a high impedance state via the HALT signal. Mileage clearly varies when it comes to hot 6502 behaviour, but in my case it looks like the CPU's ability to produce a reliable clock was inhibited at high temperatures.

[pirx]

not really on topic, but this is something I wanted to share:

http://hackaday.com/2016/07/02/the-dual-core-arm-powered-commodore-64/

[576XE]

Hi there, friends!

 

Anyway, the Idea of glueing a heatsink on some parts of Atari is absolutely ingenious.

 

No one knows why Atari memory chips become dead once upon a time...

...

[Mark loves Stella]

Hello all,

 

I just received my Rapidus and 800XL Rapidus adapter last week. But, I haven't had any luck getting it to work. I installed it on an Atari 800XL with U1mb, VBXE, and Stereo upgrade. With no soldering, it was the easiest to install, but also the most expensive upgrade.

 

So far, I haven't been able to get in to the Rapidus BIOS setup to configure it. I read in another thread that must you press the inverse button and reset to get to the menu. However, that has no effect. As a test, I removed the U1mb upgrade, just in case it was somehow defaulting to the Ultimate Setup BIOS. It will only boot to BASIC. Inverse plus reset still does not bring up the Rapidus BIOS. With the 1mb back in, the BIOS is only picking up the 6502c processor.

 

Does anyone have any ideas?

Any help would be greatly appreciated.

[tebe]

Pasiu fix my problem with Rapidus + KMK+ (Atari OS hangs when turned on), and fix problem with U1MB + SIDE + Rapidus (broken file system during writing)

 

how?

 

added some delay signal

 

Edited April 8, 2017 by tebe

[Mark loves Stella]

Thanks for your feedback tebe. I didn't get a chance to install the signal delay before my 800XL died.

I'm afraid I might put the Rapidus away somewhere that I can't find it.

 

Since I received it, I have installed it in a 600XL and two 800XLs.

None of them work after using the Rapidus on them for a time. They eventually just turn on to a blank white or blue screen.

 

I will be ordering another 800xl motherboard from BestElectronics shortly. But, I think I will just stick with the VBXE, Stereo and Ult 1MB upgrades for now. Perhaps I will be purchasing the Veronica down the road.

 

I don't know what it is with the Rapidus. But, it has somehow damaged 3 of my XLs.

[drac030]

For sure.

 

And for something a bit different. I have recently discovered that the file version of the Pitfall II works on 6502 by pure accident, because of the initializing code being seriously flawed. On 65C816 the garbage the CPU is encountering is interpreted differently, so the condition for the game to get started is a bit more difficult to meet. If anyone has a problem with that, here is a fixed version. It should work without a problem either on 6502 and on 65C816.

 

http://drac030.krap.pl/pitfall_2_fixed.zip

Edited April 12, 2017 by drac030

[lotharek]

I have noticed, that some cpus - like FJC said- are overheating

 

for last 3-4 months (after FJS`s hammering ) i test all cpus / atari i have around me...and all that led me to conclusion:

 

MEXICO CPU ...works at first boot;after few minutes of work and manual restart: rapidus core loading bar stops..

NCR - best ones ! no problems with hangs during boot. no problems at all.

 

others...hmmm.. in between

 

what is also interesting, less problematic machines are... 800XL withou freddy chip; with those, mexico cpu works flawlessly

 

tested: approx 10 cpus, 10 ataris in every combinations. took few months but... if You have xe machine - go for NCR, if You have xl machine, go for ncr in case of problems.

Edited June 8, 2017 by lotharek