SUBJ: 1764 REU & Turbo Master (R10) FROM: SSchnedler 12/08/90 S#: 508441 As a surpsing (to me) development, Paul Bosacki has a Commodore 1764 REU working with a Turbo Master CPU processor in the GEOS environment. He has described how he has done it. It strikes me as a flash of genius, although the required adapter circuit is almost trivial. We are very seriously considering manufacturing the required adapter. (The adapter also should improve the GEORAM situation which arises because not all C64s work properly with a cartridge expansion motherboard, whatever the cartridge is.) Please tell me how much interest would there be in such an adapter? Understand, it would only be for GEOS. ---------------------------------------- SUBJ: TM & RAMLINK/DRIVE (R1) FROM: Chuck 54 12/09/90 S#: 486478 How about a few words on the compatibility of Turbo Master CPU and CMD's Ramlink/Ramdrive. I'm interested in assembling a 'Killer C-64' BBS system and these are two products that would greatly enhance a BBS system on the C-64. Thanks. -::- SUBJ: We don't know yet FROM: SSchnedler 12/09/90 S#: 472312 At this point we don't know how to use a Turbo Master CPU at the same time as either Ramlink or Ramdrive. However, we are working on it, and there is reason for hope. Compatibility with RamDrive is much more likely than with RamLink. ---------------------------------------- SUBJ: TM & geoRam won't work for (R2) FROM: BrunoK 12/16/90 S#: 798661 I've tried to get the above two to work together via a Skyles board and an Aprospand board without much success. I've managed to get the combo to work for under 5 minutes before I lost control of the screen (with the skyles board). Usually I can't even get to the desktop (I'm only trying with geos) before the system locks up. I use 2 128's daily in two differentt locations and I've had the same results with each of them. While I'm disappointed, I'll say this...the 5 minutes the two were working together were nothing short of AWESOME! (btw, I was using a 2 meg geoRam upgraded by CMDR FIXER here on Q) Regards and best wishes, Bruno -::- SUBJ: Help on the way. (R) FROM: SSchnedler 12/16/90 S#:467512 This customer's problem is not unique. A number of C64s are unable to drive much of a load on the expansion bus, and some will not work with ANY cartridge on a cartridge expansion motherboard, let alone one as complex as Turbo Master CPU. The basic problem is that the cartridge expansion motherboard, and a second cartridge, add capacitance to the processor bus, resulting in distortion of the critical timing signals. The solution which we are preparing to market is a small, two-slot expansion board which is BUFFERED, i.e., critical signals from the C64 will be amplified immediately as they emerge from the computer. Likely the slots above one another in the final mechanical configuration so that the two cartridges can be in their normal horizontal positions one above the other. As a bonus, a Commodore 17xx REU will also work with proper drivers, which will be for GEOS only. As BrunoK suggests, GEOS at 4MHz with RAM expansion is impressive indeed. In the meantine, here are temporary solutions which have worked in some cases: (1) A C128 is much better than a C64 in this regard, because the C128 has more internal buffering and thus stronger drive signals sent to the expansion port. (2) One customer has added a 74LS244 buffer inside his C64 to buffer the phase-2 clock signal which is on Pin "E". (3) Sometimes an additional cartridge plugged into the end cartridge expander slot, BUT TURNED OFF, will improve the situation by acting as a bus terminator which reduces "ringing". -::- SUBJ: I have seen this too. FROM: RaymondD2 12/17/90 S#: 843650 I upgrade a lot of REU's. I test them with my Skyles 2+1 board. Most REU's work good. But all most all will at one time or a nother come back with an error. But I never get an error with the REU pluged right in the computer. Some REU's will not even work at all. I don't think it's the computer it's the REU and the expanshion board. Your board sound super for two cartridges to go in on top of the other, right? I was thinking some one my do this or have one all ready so when I upgrade the REU's up to 2 megs, I all ways put the switch and LED's on the left side of the unit. Most every one else puts then on top of the unit and this way you could not put a cartridge on top of it or any other unit that works with the REU. I will be looking for you cartridge expander if it has it so you can plug 2 on the top of each! Thanks. If you ever get your terbo master to work with the REU please let me know. I would get one right away then! ---------------------------------------- SUBJ: Master Adapter News (R1) FROM: SSchnedler 04/16/91 S#: 514664 At this time two Master Adapters are in existence. One is in the hands of Paul Bosacki for development of GEOS drivers, and the other one I have. In addition to GEOS drivers (which I hope is definite), a basic form of RAMDOS support is a possiblity. The hardware works as expected. The Commodore REU and Turbo Master communicate with each other, and GEORAM works more reliably. All that is needed is the software driver for Commodore REU. The GEOS 2.0r software already has the necessary driver for GEORAM. Master Adapter has a set of jumper blocks which are set up to configure for either GEORAM or Commodore REU. Physically there is a very short 1-3/4 inch horizontal board that plugs into the C64, with a single 74LS244 buffer IC to buffer Phase 2 clock and DOT CLOCK, and to gate R/W running back to the C64. This plugs at right angles into a 3-1/2 high vertical board that has two vertically stacked expansion slots that extend horizontally. For use with a Commodore REU, the REU is plugged into the bottom slot and rests on the table normally, except it's 1-3/4 inches farther back. Turbo Master sits dirrectly on top of the REU, and plugs into the top slot. For GEORAM use, Turbo Master plugs into the bottom slot, and GEORAM is on top. Since GEORAM is narrower, the switches on Turbo Master are still accessible. The vertical board has a 2K x 8 static RAM chip, which is the smallest convenient size, as only 256 bytes are used. For REU adapter configuration, these 256 bytes are mapped into memory at $DExx, and are visible to both the Turbo Master and the REU. All data exchange between the two is through this memory, which acts like dual port memory. As it turns out, it is not feasible to post a schematic for anyone to just build one. Physically it's cumbersome without laying out a circuit board. I tried to make a wired prototype, but it got so messy that I just went directly to circuit boards. More significantly, it turned out that one of the PAL ICs in the Turbo Master has to be replaced with a reprogrammed one, and a single jumper run between the Master Adapter and the new PAL. All of this will be supplied of course with Master Adapters. Regarding availability and cost, this is again a highly active project, and should, as they say, be ready REAL SOON NOW. We're going to try to price it at $49. At that price the Master Adapter will not itself be a profit item, but more in the nature of support for Turbo Master. We'll order production boards immediately when the Commodore REU support software for GEOS is completed. I'm not willing to risk having a quantity of Master Adapter boards made without proven software drivers in hand, since there is only one person I know who can write those drivers, and that person is not me. ---------------------------------------- SUBJ: GEOS Patch News FROM: SSchnedler 04/16/91 S#: 508863 The message "check it out" is interesting, and illustrates that there do exist better places to put the patches so as to avoid conflicts. For example, Paul Bosacki tell me that $CFE0 through $CFFF is free, but only in GEOS 2.0r. As you may know, Paul Bosacki wrote the "Geos Gateway" for Creative Micro Designs, as well as their super-fast GEOS drivers for their Hard Drives. In the process he has learned quite a bit, so much he views writing the driver for Commodore REU support through Master Adapter to be rather simple (although I don't see it that way). Also, Paul came up with the basic principle of the Master Adapter. Also interesting are available utilities like GEOS "Make Boot" which allow you to capture a patched version of GEOS in memory to make a custom boot disk. Two clear advantages. (1) The patched version loads fast, and will load at the SOFT 4 MHz switch setting; and (2) All necessary patches, like the basic "Patch GEOS" patch for Turbo Master can be incorporated into the customized captured version with no further manipulation needed. The whole process of booting GEOS is greatly simplified. So through Paul's efforts, we hope to shortly have: 1) Improved "Patch GEOS" patches to avoid memory conflicts. Also, by including this within the range of the GEOS code itself, a patched version can be saved using "Make Boot" procedures. 2) Commodore REU driver support for Master Adapter. 3) A program like GEOS "Make Boot" specifically optimized for Turbo Master CPU. ---------------------------------------- SUBJ: Turbo/Master Adapter FROM: SSchnedler 07/11/91 S#: 462915 Note: Master Adapter allows a Commodore 17xx REU to work with a Turbo Master CPU 4 MHz C64 processor accelerator when running under GEOS. Update on availability: Hardware design is ready, but will not be committed to production until software driver for GEOS is finished. Paul Bosacki, who is writing it, reports it's not that long and is almost finished; only problem is it doesn't work -- yet. Seriously, it's very close to working, and when it's ready it should really be ready. (This has been a "spare time" project for Paul, as he is mainly working on GEOS gateway for CMD. It is his knowledge of the inner workings of GEOS which makes the driver even possible, and his imagination which conceived the Master Adapter and its principle of operaton in the first place.) ---------------------------------------- SUBJ: I'll be following up FROM: SSchnedler 07/28/91 S#: 430098 with the programmer this week. I've given up predicting when he'll have it working. The GEOS driver project for Master Adapter is going alongside the Geos Gateway project for CMD, and would be finished by now if only it worked. Thus there is still some, hopefully small, problem to be discoverd and fixed. ---------------------------------------- SUBJ: master adapter reset? (R1) FROM: SIDist 08/22/91 S#: 23649 Will Master Adapter have a reset button? I love everything about my new TurboMaster, but it would be nice to have a reset button that would initialize the disk drives as well (for those fatal GEOS crashes). The new turbo boot for GEOS by Mr. Bosacki really works well. If you have JiffyDOS and TurboMaster, GEOS comes up in no time at all! -::- SUBJ: RESET Buttons FROM: SSchnedler 08/22/91 S#: 31351 Master Adapter will not have its own reset button because Turbo Master already has a reset button which resets the C64. You are reporting that your disk drives do not reset when you use the RESET button. The explanation is that some C64s send a reset on the serial bus reset line when the internal reset line is pulled low, and others do not. Yours is apparently one which does not have the serial bus reset line internally passed through. This is one of the many ways in which different C64s differ. We have a C64 here which behaves just like yours, and it is indeed a pain at times. The only remedy for you is to add a reset button to your C64 itself, and ensure it is wired to the reset line on the serial bus. I'm glad you've found and are using Paul Bosacki's MakeTurbo. It really is amazing what a competent and knowledgeable programmer can accomplish. Expect a similarly slick job on his GEOS driver for Master Adapter and Commodore 17xx REUs. ---------------------------------------- SUBJ: Master Adapter Closer FROM: SSchnedler 08/22/91 S#: 43308 Paul Bosacki reports he has completed the GEOS driver for Commodore 17xx REU compatibility with Turbo Master CPU. See text area in path: Product Information Master Adapter for more details. The driver works with any size Commodore REU. 128K, 256K, 512K, or even 1 Meg and 2 Meg expanded REUs! The driver works in conjunction with the existing CONFIGURE program (CONFIGURE 2.1 in the case of 1 and 2 Meg REUs). Also, Paul Bosacki has made sure that the REU driver for Master Adapter is compatible with his GEOS GateWay, sold by Creative Micro Designs. On a separate subject, we've reorganized the Common Questions & Answers area so there is just one question for each individual text, with headings as a sort of index. Added new item "Operation on C128" and updated "CMD Compatibility?" to mention SWIFTLink. ---------------------------------------- MASTER ADAPTER Status update 8/22/91: Programmer Paul Bosacki reports he has finished the GEOS driver for Commodore 17xx REUs, and he will send it to us by the end of August, after final testing to satisfy himself. Based on Paul's report, we are proceeding to order a production run of circuit boards, even though the disk mailed from Canada will take a while, and we will not ourselves be testing the software prior to ordering the boards. We will shortly post an expected ship date for Master Adapter, but we do not want to commit to a ship date until all the parts are ordered, with definite commitments from our suppliers. In other words, once we give a date, we hope to stand by it. We still are not accepting advance orders. Paul's driver works in cooperation with the existing CONFIGURE program, and so will work with any size REU. This means 128K 1700, 256 K 1764, 512K 1750, and even 1 or 2 Meg expanded REUs operating under CONFIGURE 2.1! It works with GEOS Gateway, which he also wrote. Based on discusions with Paul he has done a first class job, putting in a number of "extras" to satisfy himself. Master Adapter is a special two-slot expansion board. Its main purpose is to achieve compatibility between Commdore 17xx REUs and the Turbo Master CPU accelerator when running under GEOS. However, Master Adapter also improves operation with GEORAM. (With some C64s, operation of GEORAM and Turbo Master is unsatisfactory when using an Aprospand expansion mother board.) Master Adapter has not yet been released as a product. We will keep current status information posted here, as well as on the message board. We anticipate the price will be $49, even if there is minimal profit, justified on the basis it will support the basic accelerator. Advance orders are NOT being accepted. Paul Bosacki, who wrote GEOS Gateway for Creative Micro Designs, is also doing the necessary REU driver for Master Adapter. The Master Adapter software is compatible with GEOS Gateway. Also, credit to Paul Bosacki for conceiving of the principle of operation. Previously we had always thought Turbo Master CPU and Commodore REUs were hopelessly incompatible, but Paul figured out a way. -::- Physically Master Adapter is two small circuit boards joined in a "T" configuration. A small board comes horizontally out of the computer, and has 74LS244 buffer IC on it. The small board plugs into a larger vertical board with two expansion sockets that are stacked vertically and oriented horizontally. This allows the REU plugged into the lower slot to sit flat on the table in almost its normal position, with Turbo Master in the upper slot directly on top of the REU. Thus both cartridges are horizontal, for a reasonable physical configuration. With GEORAM, Turbo Master plugs into the lower slot, and GEORAM into the upper slot. Electrically, the vertical board includes a TTL logic IC, configuration jumpers, and a 256 byte static RAM which is visible in memory at $DE00 to both the accelerator and the REU. The two cartridges communicate through this static RAM. Normally the accelerator sees this 256 byte RAM. But when the REU requests DMA, Master Adapter tells Turbo Master to relinquish the bus, and REU has access. While it may seem cumbersome to transfer all data through a single page of memory, at computer speeds (particularly at 4 MHz) it makes no difference. Actually this is very much the way GEORAM works. For GEORAM mode, the 256 byte RAM is not active. Operation is improved compared to an ordinary cartridge expander due to buffering and due to the small physical size to minimize capacitive loading. ---------------------------------------- SUBJ: Master Adapter (R1) FROM: Milwaukee1 09/12/91 S#: 505612 With Master Adapter installed, will I still be able to use the REU with non-GEOS software when TurboMaster is disabled? (Or will I have to remove the entire assembly and plug in just the REU?) Will you be releasing information for programmers regarding Master Adapter? -::- SUBJ: Probably and Yes FROM: SSchnedler 09/12/91 S#: 819451 I've just looked at the schematic, and it appears that with a simple jumper clip the REU can have its normal access to the C64 bus while the Turbo Master CPU accelerator is disabled. That's a very good question you asked. I'll check it out for sure, and add it to the documentation if it works. There's no secret to the programming. Briefly, the REU registers at $DFxx are available and work normally. However all data transfers to and from the REU and the RAM in Turbo Master go through a 256-byte RAM at $DExx. So to transfer from C64 to REU, your program must first write a block to $DExx, and then tell the REU to transfer from $DExx to wherever in the REU you want it to go. From REU to C64, the process is reversed. In short, there is one page of RAM at $DExx visible to both devices. The Master Adapter has this RAM on it, as well as logic to keep the two devices from conflicting. Normally the Turbo Master has the bus and access to the 256-byte RAM. However, when REU pulls DMA low, Turbo gets out of the way. (The REU DMA line is routed through a gate to the Turbo to simulate a VIC chip BA access, rather than to the C64. The C64 DMA line is always pulled low while Turbo is enabled.) ---------------------------------------- SUBJ: Master Adapter (R1) FROM: ShadowM 09/29/91 S#: 46974 What's the latest on this long-awaited device? -::- SUBJ: The latest is ... FROM: SSchnedler 09/30/91 S#: 46989 Our board supplier has promised delivery October 16th. Other parts are currently arriving and should all be here by then. In the absence of some disaster, we should have a few Master Adapters assembled, tested and ready for shipping Monday October 21. We will be hand-soldering these, and production rate is unknown. The necessary driver program for Commodore 17xx ReUs has still not arrived from Paul Bosacki in Canada. Although the postal strike there is over, the package he sent seems to be lost, and so he recently sent another one. ---------------------------------------- MASTER ADAPTER Status update 10/6/91: Our board supplier tells us to expect delivery of bare boards by Wednesday October 16. Other parts have been arriving. In the absence of disaster, we expect to be shipping some by Monday October 21. The first batch is being hand assembled and soldered, and we do not know how fast we will be able to make them. Our ad on page 32 of the Nov/Dec 1991 issue of RUN which is just out says "It's Here!". Unfortunately, that's not quite right, but was our best guess when the ad was placed, well in advance. A potential problem is that we STILL do not have the necessary driver program to operate the REU under GEOS, even though programmer Paul Bosacki has sent us two packages. The postal and other strikes in Canada are causing a lot of problems. We are confident that if Paul says his driver works, it does. But we would feel better having it in hand. We will have final pricing this week. Some parts are costing more than we would like, and we may increase the price to $59 assembled, with Master Adapter in kit form at $39. Assembly involves soldering four 44-pin connectors, a 24-pin RAM socket, two other IC sockets, a 12-position jumper block, and a few resistors and capacitors. A kit is feasible since we plan to hand assemble and solder the first production run. Also, the circuit is simple enough that most will work first try. Advance orders still are not accepted. At these prices there actually is minimal profit for us, and a lot of trouble. However we are continuing with the project because it is started, and we like to support the basic accelerator. Based on running GEOS at 4 MHz with RAM expansion, I can say the difference is amazing. I understand now why so many people wanted REU compatibility when Turbo Master first appeared. Paul's driver works in cooperation with the existing CONFIGURE program, and so will work with any size REU. This means 128K 1700, 256K 1764, 512K 1750, and even 1 or 2 Meg expanded REUs operating under CONFIGURE 2.1! It works with GEOS Gateway, which he also wrote. Based on discussions with Paul he has done a first class job, putting in a number of "extras" to satisfy himself. ---------------------------------------- SUBJ: Master Adapter Almost Ready(R1) FROM: SSchnedler 10/28/91 S#: 498247 We have the boards and components, and are currently assembling them. Manual is in draft form. Driver program still is not here. Was sent again (for the fourth time) by Federal Express on Saturday 10/26, and with luck it will be here 10/29 or 10/30 depending on customs delay from Canada. Hopefully the disk will not be zapped by X-rays. Paul Bosacki called me this evening to report he has found a bug in his driver program which causes an incompatibility with geoCalc and geoChart. He thinks he knows what it is, and will send a replacement soon. In the meantime, we will release the product with his original version, hopefully in a matter of days, and send later send updates to initial customers. Q-Link text area will be updated when we are actually ready to ship. ---------------------------------------- SUBJ: Master Adapter Problems FROM: SSchnedler 11/03/91 S#: 480775 It's still coming, but not easily. Please see status update in text area posted this date -- Product Information Master Adapter The stress level is running a bit high here lately. We have a backlog of orders. no credit cards will be charged until actual shipment, and we are not going to ship until this thing is right. -::- ---------------------------------------- MASTER ADAPTER Status update 11/3/91: We're having problems, but this product WILL DEFINITELY be released soon. We have the boards, parts and driver program from Paul Bosacki all in hand, and should theoretically be shipping, but a disaster of sorts has struck. In short, we've run into another situation where different C64s are different. The Master Adapter hardware works on some, but not others. Fortunately there is a very straightforward circuit modification which should approach 100% effectiveness. Unfortunately the circuit boards are already made, and so we will be hand-modifying each one of the initial batch. The previously-unsuspected hardware problem led Paul Bosacki to believe he had a software problem which he proceeded to "fix" but then was unable to test, thinking his prototype Master Adapter was broken, but actually was using it on a different computer. Unfortunately the "fix" did more harm than good, as often happens. But now with the hardware working, final testing of the software should proceed without difficulty. By the way, we have found that Federal Express does deliver from Canada to the U.S. in one or two days, and that is the method we are using in the future to get software from Paul Bosacki. Initial pricing will be the $49 advertised, simply because that's what we advertised. However, price is likely to increase to $59 after the initial batch are shipped. At that time we may offer Master Adapter in kit form at a lower price. At these prices there actually is minimal profit for us, and a lot of trouble. However we are continuing with the project because it is started, and we like to support the basic accelerator. Based on running GEOS at 4 MHz with RAM expansion, I can say the difference is amazing. I understand now why so many people wanted REU compatibility when Turbo Master first appeared. Paul's driver works in cooperation with the existing CONFIGURE program, and so will work with any size REU. This means 128K 1700, 256K 1764, 512K 1750, and even 1 or 2 Meg expanded REUs operating under CONFIGURE 2.1! It works with GEOS Gateway, which he also wrote. . ---------------------------------------- SUBJ: Bosacki reports success (R2) FROM: SSchnedler 11/28/91 S#: 45336 I spoke this evening (Thursday) with Paul Bosacki who reports he has found and fixed the problem with his REU driver for Master Adapter. He could have sent it by Federal Express today, but would rather spend a bit more time testing this evening. He plans to send a disk by Federal Express on Friday. It may arrive here Saturday. If so, and if our tests are successful, we will begin shipping Master Adapters to customers with REUs on Monday. I was beginning to think a lot of effort and money was down the drain in designing the Master Adapter board, and doing a production run. There's hope yet. -::- SUBJ: Not Saturday (R) FROM: SSchnedler 11/30/91 S#: 11656 While a disk with Paul Bosacki's REU driver did not arrive Saturday, it was sent, and should be here Monday. I spoke with Paul today, who says he will be "very surprised" if I find it does not work. ---------------------------------------- SUBJ: Master Adapter Ready (R1) FROM: SSchnedler 12/07/91 S#: 444216 It looks like we actually will be shipping Master Adapters with the REU driver beginning Monday December 9th. More problems developed and were solved since my last posting a week ago. Delays in Federal Express due to error in our account number and then misdelivery led to our not receiving the disk until Wednesday, and then it still did not work. Aparently there was a difference between Paul Bosacki's C128 and our C64. Paul was able to duplicate the problem on his C64, and fixed it around midnight Wednesday. Fortunately he also sent his source code, and was able to tell me what to do by telephone. So Thursday I fired up geoAssembler, which I had never before used, and after many hours successfully reassembled the REU driver program and tested it. This Saturday morning I have been testing Turbo Master with a Master Adapter and a 1764 REU under GEOS 2.0 on a plain C64. It seems to work, and so we will put the REU driver on a revised Turbo Master disk (revision date 12/9/91) and begin shipping Monday. I should add that GEOS is a demanding program, and there are some C64s with which a Turbo Master simply will not work. The added complexity of the Master Adapter certainly will not improve this situation; it can only make it worse. Therefore I fully expect, unfortunately, that there will be some people who cannot use Master Adapter. In those cases, there will be nothing further that can be done, and the Master Adapter will have to be returned for refund. This project has been a nightmare. -::- SUBJ: Thank you! FROM: ShadowM 12/07/91 S#: 43328 I am sure that all truly hard-core Commodore users appreciate the tremendous effort your company has put into this project. (Personally, I would much rather have the speed of a TurboMaster with extra memory than all the features of RAMLink.) Will Mr. Bosacki's source code be made available also? ---------------------------------------- SUBJ: Master Disaster? FROM: SSchnedler 12/14/91 S#: 50630 Since we commenced shipping this past Monday, a number so customers have reported their systems crash in various exciting ways when running GEOS in Turbo Master with a Commodore REU and Master Adapter. However, others report success and are quite pleased with the performance. (In GEORAM mode there have not been problems.) Based on calls, it may be that less than 50% are working correctly. All of this indicates some kind of serious hardware incompatibility with different C64s. We will be working on several approaches this weekend involving hardware design. Within the next several days there will either be a solution, or the entire Master Adapter project will be killed in order to cut our losses, which at this point are several thousand dollars and an amazing amount of time. ---------------------------------------- Mail From: SSchnedler Date: Friday 27-Dec-91 22:39 EST A belated answer to your E-Mail of Dec. 14. Don't run out an buy a GEORAM yet. There's still hope for the REU driver. I've posted some new stuff this date. Thanks for your patience. ---------------------------------------- MASTER ADAPTER Status update 12/27/91: Master Adapter remains available for GEORAM users. While we originally thought the REU driver was working correctly, there definitely is a problem, which Paul Bosacki acknowledges. At the same time, THERE DEFINITELY IS HOPE, and we ask for patience from everybody, especially those who have purchased a Master Adapter for REU use, and at this point are holding onto a piece of hardware that needs correct software for it to work. There is no point in a blow-by-blow report of the various things we have tried the past several weeks. We do know that data is slowly and somewhat randomly being corrupted as it transfers to and from the REU, and that the hardware is not at fault. Our current theory is that IRQs are causing the problem. Paul's next experiment will be to move a buffer from $E0xx to $E1xx, which is the buffer area used by the standard GEOS 2.0r driver for GEORAM. I spoke to Paul this evening (Friday), and he says he will be able to work on the Master Adapter driver again on Monday Dec. 30. On a slightly different subject, Paul says he has developed a driver to allow a 2 Meg CMD RamDrive to work with Turbo Master and Master Adapter under GATEWAY for GEOS (sold by Creative Micro Designs). The RamDrive driver uses the GEORAM mode of Master Adapter, and so is not subject to the REU driver problems discussed above. We do not have the RamDrive driver, certainly have not tested it, and have no further information at this point. We are viewing it merely as a possibility. Initial pricing will be the $49 advertised, simply because that's what we advertised. However, price is likely to increase to $59 after the initial batch are shipped. At that time we may offer Master Adapter in kit form at a lower price. At these prices there actually is minimal profit for us, and a lot of trouble. However we are continuing with the project because it is started, and we like to support the basic accelerator. Based on running GEOS at 4 MHz with RAM expansion, I can say the difference is amazing. I understand now why so many people wanted REU compatibility when Turbo Master first appeared. Paul's driver should work in cooperation with the existing CONFIGURE program, and so will work with any size REU. This means 128K 1700, 256K 1764, 512K 1750, and even 1 or 2 Meg expanded REUs operating under CONFIGURE 2.1! It works with GEOS Gateway, which he also wrote. . ---------------------------------------- SUBJ: M/A Ready Again? (R1) FROM: SSchnedler 01/02/92 S#: 426583 I spoke this evening with Paul Bosacki regarding progress on fixing the REU driver for Master Adapter. He believes it is now working, and has tested it on several computers. He says he would like to continue to test it over the weekend, and send it to us by FedEx on Monday. We are extremely anxious to have this driver in had, for obvious reasons. At the same time I am glad to see that Paul intends to thoroughly test it this time. IF this all works out, and with a bit of luck, Master Adapter will be underway again next week. ---------------------------------------- SUBJ: In anticipation (R2) FROM: GregoryD12 01/05/92 S#: 516355 of a working TurbREU, is it safe (hardware-wise) to leave the "optional internal jumper" inside the TurboMaster permanatley, even with the TM plugged directly into the computer? -::- SUBJ: For me (R) FROM: RandyS38 01/06/92 S#: 437188 the internal jumper has caused no problem with any software that I used before. -::- SUBJ: Design Objective ... FROM: SSchnedler 01/12/92 S#: 457476 ... was that the optional internal connection to facilitate Master Adapter would be compatible with a "standard" Turbo Master setup, so there should be no problem. The internal connection should remain pulled up to logic high when Master Adapter is not in use, which is what the /GAME line does. If we ever make another run of Turbo Masters, the board layout will be modified at that time to make the connection. ---------------------------------------- SUBJ: Bosacki Continues FROM: SSchnedler 01/12/92 S#: 457491 A week ago Paul Bosacki thought he had the REU driver for Master Adapter working perfectly. However, his further testing has revealed some subtle problems. He is still working this evening to discover the cause or causes. I have been speaking to Paul by telephone almost every day for weeks now, and every day I hope the news will be good. If he gives up, that will be the end of the Master Adapter project; I don't have the slightest idea who else could do it. Fortunately, Paul seems to have made this a personal mission to stick with the problem until it is solved. He could very easily just decide to tell me he can't do it. ---------------------------------------- SUBJ: The good news is ... (R4) FROM: SSchnedler 01/16/92 S#: 473249 ... Paul Bosacki has not yet given up on the Master Adapter REU driver, but of course it's not yet working. I have spoken with him nearly every evening for weeks. Tonight he tells me that he is starting over to recode the driver, taking a slightly different approach. The code is not unduly long; about six pages of assembly language. It is just rather complex in the things it is doing to GEOS. There appears to be some subtle problem in the existing code that repeated and careful study hasn't revealed. A difficulty in getting it to work is that the operation is hardware intensive AND is in the GEOS environment; as a result, conventional software debugging techniques are of little help. He has observed corrupted data following a transfer, and knows generally when it occurs, but not why. Our repeated and various tests lead strongly to the conclusion that we are now dealing with a sotware problem, not a Master Adapter hardware problem. -::- SUBJ: Good news (R) FROM: GaylnH2747 01/27/92 S#: 794298 Hi Steve, just making my weekly check on master adapter progress. Dusted off my C-64 (classes started two weeks ago!!) and put in a ten hour session building a newsletter. WOW!! I have to say again GEOS with Turbo Master is amazing. The DOS machines at school seem primitive and even my bosses 386 running Ensemble isn't really any faster. (PC/Geos does have some neat features though). What I am saying is THANK YOU -- Turbo Master and Master Adapter used with an expanded Georam make GEOS a real joy to use. TM evens speeds up printing -- a 12 page newsletter (w/geopub) printed in 21 min. compared to 45 min. without TM. Please keep hunting for those bugs, I am sure the many REU equiped GEOS users will be amazed with Turbo Master boosted GEOS. Good luck and Thanks again!! :) Gayln T. Hall ---------------------------------------- SUBJ: At the risk ... FROM: SSchnedler 01/30/92 S#: 468424 ... of having hopes dashed again, Paul Bosacki reports he has had the new REU driver for Master Adapter working since last Saturday, but did not tell me due to the many false alarms in the past. He says he has tested it extensively, and it runs with all GEOS software he has tried. Current plans are for him to send it to me by FedEx Friday January 31, to arrive here Monday February 3. I will test it thoroughly and, if it works, we will send new disks to all customers who already have Master Adapter, and then have a look at the back order situation. As a brief explanation of the nature of the bug, the Master Adapter hardware only permits REU data transfers to and from a 256-byte area of memory at $DExx. Pointers are used to indicate to the GEOS REU routines what address in the C64 is to be the target of an REU transfer, and it seems there were circumstances when the pointer would point to other than the $DExx block. Attempted transfers to or from other areas of memory resulted in a seemingly random corruption of memory, which resulted in a crash sooner or later, usually sooner. The driver was rewritten with a view toward guarding against such out-of-bounds data transfers. Hopefully I will have confirmation posted early next week. Thanks again to all for patience. ---------------------------------------- SUBJ: No news is ... FROM: SSchnedler 02/12/92 S#: 40642 ... generally bad news, as I tend to report only positive things regarding the Master Adapter project. My silence over the last two weeks means troubles, but there continues to be hope for the Master Adapter project. Paul Bosacki's withdrew his report of all problems fixed after trying the REU driver on a C64. He actually was confidently demonstrating the system for a frient. His previous successful tests had been on a C128. He did send me a disk with the driver program and source code in its defective state, and I am now deeply deeply involved in trying to debug it. Tonight, assisted by telephone discussions with Paul, we found a definite error in the driver program. Whether it is the only error remains to be seen. We have been at this point many times; that is finding an error and being convinced it is the last one, only to find out problems still remain. I will have the driver program reassembled and tested again in a day or two. Who knows. The current error in the program relates to sprites. They must be turned OFF for an error-free REU transfer when using Master Adapter, and we found a place in the driver program where a REU transfer is done without turning OFF the sprites. Those of you patiently waiting with Master Adapters in hand can demonstrate this for yourself using the simple TEST REU program on the disk. From Basic, turn ON the sprites by entering POKE 53270,1. Likely no sprites will be visible, but they willbe enabled. Now run the TEST REU program with Master Adapter assembled, and likely there will be errors. Turn the sprites back OFF with POKE 53270,0. ---------------------------------------- SUBJ: Master Adapter REALLY ready(R2) FROM: SSchnedler 02/17/92 S#: 436079 The REU driver program for Master Adapter has successfully been debugged, and appears to be working very well. This week we will be sending new disks to all customers who have Master Adapter, and we will then deal with the back order situation on the Master Adapter hardware. To make a long story short, in retrospect the solutions seem to be simple. First, Paul Bosacki found a few weak parts in his program, like wild pointers, but actually the code was pretty good. Paul was telling me the problem must be hardware, while I was telling him, no, it must be the programing. The truth lay somewhere in between. Paul was at a dead end as far as he was concerned, which is why he sent me the non-working source code two weeks ago for me to look at. It turns out that sprites must be disabled for an error-free REU transfer through the Master Adapter. So we had to ensure the program code always turned off the sprites. In addition, for reasons unknown, the REU SWAP command is very unreliable with the Naster Adapter, even with the sprites turned off, even though STASH and FETCH appear to be 100%. This is something I determined yesterday using separate trest programs. So all use of the SWAP routine was eliminated. This is the final thing I changed this afternoon, and tested extensively this evening. So, there were hardware-related problems which could be worked around with programming. Paul and I have discussed this at length, and neither of us have any idea why REU transfers through the Master Adapter fail to work with the sprites enabled, nor why SWAP doesn't work al all. Fortunately, GEOS itself never attempts to use the REU SWAP command. Incidentally, with Paul Bosacki's permission, the source code for the Master Adapter REU driver will be on the disk. ---------------------------------------- SUBJ: Disks in Mail FROM: SSchnedler 02/19/92 S#: 398150 Disks were mailed today to all customers who already have a Master Adapter an who have been unable to use it with the Commodore 17xx REU. The "fixed" driver program is on the disk, and the disk revision date is 2/18/92. If your unit successfully runs the "TEST REU" program on the earlier disk dated 12/9/91, the probability is very high it will now work with the 17xx REU under GEOS. We will now address the back order situation for the Master Adapter hardware. ---------------------------------------- SUBJ: Master Adapter wow! (R1) FROM: ShadowM 02/28/92 S#: 843939 I finally have the working Master Adapter REU driver, and it survived everything I threw at it. Especially impressive was switching back and forth between preview and zoom in a complex geoPublish document. I have two questions, though: 1) Do you or does some enterprising person out there plan to market an enclosure for the Master Adapter? All those exposed IC's make me nervous. And if so, would it be possible to have a switch for those four jumpers to go from Master Adapter mode to REU mode? 2) I find that certain keys don't work in geoDebugger when in Master Adapter mode, unless I switch down to 1MHz. Has anyone else experienced this? Any ideas? Thanks for your perserverance and indomitable spirit Mr. Schnedler, and thanks for Mr. Bosacki's source code also! I like how it hides out in upper memory 'til it's activated (sort of like the Michelangelo virus). ;> -::- SUBJ: Answers FROM: SSchnedler 03/03/92 S#: 816451 1) No enclosure is planned. 2) It is not surprising that geoDegugger does not work at 4 MHz due to the nature of a debugger. I am glad that it works at all. Glad to hear yours is finally working. ---------------------------------------- SUBJ: RBOOT (R1) FROM: ShadowM 04/02/92 S#: 408938 What would it take to patch RBOOT for Master Adapter? You would have to use code like that in the Master Adapter patch to pull the reboot code down to $6000 from the REU, and then patch the code at $6216 that pulls down the rest of the Kernal in pieces, right? What I can't seem to find is where the main part of the Kernal above $CFFF is pulled back down from the REU. It isn't just assumed to still be 'under' the ROMs, is it? Also, I don't understand why the reboot code switches from JSR $6216 to JSR FetchRAM where it does. Can anybody help? -::- SUBJ: It switches from FROM: NateF 04/03/92 S#: 518310 JSR $6216 to JSR FetchRAM because until it loads in the GEOS Kernal, there is no FetchRAM routine available. As for the "under the ROM's" assumption, no. It is all loaded in from the REU. Nate -::- SUBJ: Master Adapter/RBOOT (R1) FROM: ShadowM 06/16/92 S#: 835155 I just uploaded a M.A. version of RBOOT to the User Applications area in CSS/GEOS Arena. I'd appreciate any bug reports, especially if it runs on some systems and not on others! The file name is "ma-rboot.sda". -::- SUBJ: ma-rboot FROM: Tiger17 06/23/92 S#: 405423 If you really did this you are wonderful! I'll get it right away and give it a try. If it works it should be here in Schnedler's library. tiger ---------------------------------------- SUBJ: Been nice (R4) FROM: Dauntless 06/04/92 S#: 442993 doing business with you. Hope you have a good future. -::- SUBJ: It's... (R) FROM: Zimbo B1 06/07/92 S#: 467598 a miracle that the C-64 has survived as long as it has, thanks in part to some very good programs that exist for it. All the same, I don't think it'd be a good idea anymore for first time computer users to pick a Commodore 64... ---Joel Kolstad -::- SUBJ: RBOOT? (R) FROM: Tiger17 06/08/92 S#: 466866 RBOOT for GEOS with Turbo Master and Master Adapter? Really? Best of luck. Lack of an RBOOT is a big drawback to this otherwise great combination. Please keep us posted on your progress. - tiger -::- SUBJ: I started on the 64... FROM: Ski crazy 06/08/92 S#: 416291 a little more than a year ago and I'm not a bit sorry. It's a great machine and does everything I need. I expect to be using it many more years.:) -Tim Mail From: Tiger17 Date: Tuesday 23-Jun-92 00:57 EDT Thanks for your work. I just looked in the library and ma-rboot hasn't been made live yet. I'm very anxious to try it out since this was a major short- coming of the TurboMaster/Master Adapter combo. ---------------------------------------- SUBJ: So long ... (R2) FROM: SSchnedler 06/30/92 S#: 450057 Effective this date, formal Schnedler Systems product support on q-link ends, as we are entering other markets. We remain available by mail and telephone. Programmer's utilities and control interface boards remain available. -::- SUBJ: A dark day... (R) FROM: GregoryD12 07/01/92 S#: 804577 but many thanks for your help and support. Best wishes on your new areas.. -::- SUBJ: Yes, a sad day indeed.... FROM: Dream DOC 07/01/92 S#: 401351 but, indeed good luck in all of your future endeavors. Wonderful potential was never fully explored!!! Thanks again for what you offered. :D SUBJ: Leaving so soon? FROM: Breadstick 07/03/92 S#: 819329 I feel it is a great loss for SS to be ending work on the Commodore. I know I tried buying a TM but was told that they were all gone and may not be reissuing them. If they could make it compatible with a RamLink, I know I would probably pay up to $180 for it. Am I alone here, or is this message base being totally abandoned? I hate to see it when we lose one of the only 2 manufacturers for new Commodore products... I guess this will just put all the pressure back on CMD for development.