tf_hh

Thanks :-) No, such special sockets I never find, or they´re sold out or you´ve to buy 1000 pieces for a horrible price. The PCB´s thickness is reduced to 0.8mm instead of the usual 1.6mm and the via´s for the sockets I´ve made bigger in the Eagle library (bigger holes drilled). So any standard machine-head precision socket will fit completely through the holes and when no parts at the bottom side are placed, it fits well and strong.

Yes, the special version for the 600XL has two SRAM chips onboard. And the new version is small enough, so the removal of the RF modulator is not needed anymore.

Yes. Main schematics and outline (shape) for the new PCB is already done. But there´s something more to do until I can order the first prototype for testing. Unfortunately workload in the paid job is very high, so less time for Atari projects. And currently I´ve too much irons in the fire

Hi, just want to announce that I removed the SRAM 512 KB memory expansions V4.2 and V4.3 from my active projects. You can see the always updated infos about my projects here (be informed: within the next 2-3 weeks my domain will change to another ISP, so there may be a few days of disturbed service of van-radecke.de). A new version of my classic SRAM memory expansion is made, the version 4.4: This new version is smaller than the both versions before and fits now in ALL Atari XL, XE and XEGS computers - except the Atari 600XL. For the 600XL a special version is WIP, the 576 KB SRAM memory expansion for 600XL. This new version includes an update to 64 KByte main (base) memory (old DRAM will be removed) and 512 KB expanded memory like usual. Please use the link above, the linked PDF file will be constantly updated and includes all infos you need! Jurgen

BTW... some weeks ago I´ve to test the serial communications for Dropcheck´s upcoming SIO2IO board, so I grab my good old US Robotics Courier Dual Standard modem and try to connect some BBS... and, I was very surprised, it works better than expected! Of course I´ve no POTS or ISDN landline anymore, here in Germany nearly all fixed-net connection are Cable or DSL/VDSL using VOIP for telephone services. So I connect the USR modem to the analogue port for POTS devices of my router and dial some numbers. Connections with 300 and 1200 bps works absolute fine, no garbage (this speeds only without MNP or V.42). Using 2400 bps some garbage appears, but it works fine. Step-by-step I increase the maximum connection speed up to 33.6k using V.42 error correction (MNP - no chance). The connections were reliable up to approx. 26.4k. At all speeds above a carrier-lost within 2-3 minutes will appear. A connection using V.32bis (14.4k) with V.42 stays stable for 30 minutes. Never thought this way possible using VOIP connections... Maybe the renaissance of BBS is possible :-)

Very nice! The 1030 is the one and only Atari XL style item I didn´t own... nearly impossible to get one in Europe. I bought two times one at eBay USA, but in both cases the seller use such a bullsh*t packaging that I got only a puzzle, not a device. Ok, paypal refunds money in both times when I sent the pictures of the parcel, but... very unsatisfying. If anyone here has one to sell and would be able to pack a good parcel... let me know :-)

Jon, I agree about 99% with all you wrote here about the situation. I also intend to say "scumbag" is too hard. Luckily (regarding loss of hardware and/or money) I never get in touch with Sloopy, so I watching the whole show from a distance. The one and only thing making me personally angry is one simple thing: COMMUNICATION. Every human beeing makes mistakes. Everybody overrates him/herself sometimes. Sometimes only bad luck is the reason for not completing promised stuff. Sometimes life is just fucked up. We all know this. Most of us have experienced bad stages in life the one or other way. But... when there are problems, issues, whatever... communicate - the truth, of course. Maybe some weeks he thought "all will be good next day", but not later than 3 months, when situation goes more worst then before, clear and true words are appropriate. If he gets jobless, struggled with debts, illness, cancer, whatever... and needs the money to keep alive... not fair for those who paid for projects or sent hardware in, but understandable from a human´s beeing view. But playing hide´n´seek and telling lies is not the way to get sympathy. Just my 2 cents...

Could you sent a high-res picture of your PCB with this FDC? And, additional wish, are you able to read out the EPROM/ROM? IMHO the 1770 can´t be used with different firmware, but I may be wrong.

Cool - what´s the interface of the cat?

1st, there´s no second ROM - the left block indicates the frist 8 KByte of the O.S. ROM area between $C000 and $DFFF (incl. Selftest mirrored from $D000-$D7FF to $5000-$57FF) and the right block shows the checksum test result of the $E000-$FFFF area (also 8 KB). The complete 16 KB are located in one chip. The BASIC ROM was not recognized by the Selftest. Because the computer boots straight into the Selftest, but shows no defect RAMs, IMHO only three options possible: - Just a bloody simple short to ground (read below) - Defect O.S. ROM - Defect MMU This order beginning with the upmost probability What I mean with the first bullet point is: Check first if only a short-circuit of one of the cartridge´s / ECI´s connector pins has contact to the metal grounding plate. I´ve seen many XE systems, where the pins aren´t shortened (cutted), specially at the ECI port (2x 7 pins). Atari didn´t insert any fishpaper between the metal plate and PCB, so if there´s one of the pins too long, it might have contact to ground in same cases. Sometimes the computer doesn´t boot, sometimes this failure happens.. So simply re-test the board without metal plate loose, when the XE boots into READY prompt, you know what to do - cut the too long pins, that´s all.

I´m pretty sure that UAV won´t work without changes to it´s circuit. The FGTIA (SECAM GTIA) has also four Luminance-lines and CSYNC (composite sync), which is good for UAV, but there´s no single color signal. The Jerzy schematics show a "composite" output, but this is composite sync IMHO, not composite video. At least the videochip (NCS2563) used at the UAV PCB doesn´t say anything of SECAM capatibility in his specs. I can make a test, if my current TV is able to display SECAM (for a positive test using the default video jack).

You must have a 48 KB memory expansion to it work. Even when the menu software would work, the loader needs more space and/or the program to be loaded needs more then 16 KB RAM.

Right, these coils are for setting up the base frequency for TV channel(s). First I would try to use "fine-tuning" option of your T.V. - if this won´t help, you can try to change the coils´ settings. Use non-metal tools for this!

- double post -

At least you should install back the modulator´s metal top cover. The TV screens looks like somebody changes the setting of the HF coils in the modulator circuit. Before you re-assemble the 600XL, remove the capacitor C109 located at the upper left corner of the 4050 chip. Most times it´s a disk capacitor, brown. Removal of this scrim diffuser will significantly enhance the picture quality overall.

The PAL computers predominantly set to use UHF with channel 36 and have no channel switch. Most PAL cases didn´t have the hole for a switch, some models have a dummy plug at the switch´s hole. Some very very rare XL models have a VHF based RF modulator using channel 3/4 - like all PAL Atari 400 or 800 (they use also only channel 3 and 4 in PAL).

Yes, I also have had a dozen or so PAL XE systems with only GTIA in a socket. This socket was installed during initial production stage (soldering bath) and NOT manually afterwards. IMHO this may be proof for the fact, that Atari engineering did know they have problems with some GTIA during the XE production times. And maybe - conspiracy mode on - Tramiel also knows that, but decide to ignore it.

Me too But these selectors are very hard to get today...

Cool - you´re offering to code a new or patched XF551 firmware or better Hyper-XF? Then please go ahead, then I´m pretty sure that Dropcheck will include this in the layout - that´s easy, as three I/O pins at the 8040/8050 are unused. Otherwise such comments are nice, but it´s always the same problem: If nobody wants to code the software/firmware, then the hardware is useless.

Yes, sometimes. IMHO two three reasons cause this issue: - Defect PAL sync color clock circuit - GTIA itself - Combining Luminance & Chrominance the dirty way like Atari did it Specially at the XE series I found relative often a defect 74LS74 (or sometimes also used: 74HCTLS74). This one is located near the BASIC ROM and and syncs the PAL color carrier frequency with the system clock. Most times it just broke, so no color appears and you get a black & white picture only. But in some cases color bleeding or wrong colors also occurs. I would strongly suggest to change it. Second trouble maker is the GTIA itself. In some rare cases there´s a timing difference betwen the color-output and luminance plus sync. This effect changes after 20 or 30 minutes of operation slightly. I saw this only at PAL GTIAs and only at the AMI ones used in the XE series - but these GTIAs are not the one with the well-known "chinese PAL GTIA issue". At last the not very well done mixing of Chroma and Luma to get composite video may be the problem. As you´re able to use S-Video, I suggest to remove the RF modulator for a test - disables composite video, but S-Video works fine.

Yes, the datasheet shows also some kind of MCU. But it contains a (mask) ROM. I´m wondering about that, Atari most times uses external ROMs/EPROMs (XF551, ST Megafile series, SH204/SH205 and many other Atari products) when standard 8-Bit CPUs like Intel´s 80xx series are used. Maybe I found a way to read out the content :-)

IMHO this is no problem - hundreds of systems very delivered with that type of DRAM. At last even 250 nS is enough. Even some late XL series were equipped with "-3" DRAMs (NEC uPD4164-3 for example) and run perfect until today. From my personal experience the MOSTEK DRAMs have another issue: They consume relative high current at the -5 volts power - more than other DRAMs of that era. Due to the small dimensioned -5 volt path at the 400/800 power board - only used for the DRAMs - and the typical aging of electrolytic capacitors I would suggest to change the six major electrolytic capacitors at the powerboard first. Most of DRAM failures can be fixed this way when the genuine memory from the late 70s is used.

Yes. The Dragon Cart II will be realized as a PBI (XL series) and ECI (XE series) card. A few months ago ol.sc starts to talk with me about the idea of a Dragon Cart II, the first attempt was also realized as a cartridge. My personal suggestion was not to make another cartridge solution due the following reasons: IMHO the main success for an Ethernet solutions rely on two worktypes: Multiplayer-Gaming and Internet communication. For gaming most users today uses Flashcards for the games. Or the game itself will be on a cart. When the Ethernet-interface would also be realized as a catridge, at last you´ve to "stack" the carts - not a very nice thing. Will raise the costs of production, more instable, and maybe the registers used on both carts will interfere with eachother. Also some external SDX solutions with not fully decoded address areas may not work fully with another solution acquiring the narrow $D5xx address range. A PBI or ECI device can be managed at the whole addressbuss, can mirror memory, can switch off partial address ranges and more. This would be more compatible, configurable and better in physical handling. I´m pretty sure, both solutions have their fans. And it should be no problem, as the same Ethernet SoC is used, new developed software should support both. Just my 2 cents, of course.

Hi, sorry for this hard necrobump... but looking for a Z80 CPU in my drawer I found two unidentified chips with this marking: MOSTEK MK3870/20 MK14679N DALLAS C015946-09 The last line looks like an ATARI partnumber and I found it in this thread in post #1, but without any explanation. I´ve two of them. Anybody an idea, for what these guys are used? Here´s a picture:

Hi Michael, I have no problem with using the nice Joy2PIC-Stik, but I´m wondering what´s so special with this one (12F1571). My GALEP-5 device supports over 500 different Microchip PIC devices, but none of 12F15xx. And also the well-known chinese programmers like TL866 etc. supports nearly all Microchip PICs, but also not this one... strange thing. I´ve ordered some of them now at OSH-Park. 1st I was planning to build it using a development breadbord, but... costs more time. Thanks for info!