** 2 page Q&A pages / 2260 words ** ** Q ** I have a Mega/STe and a SLM605 laser printer with SLMC804 laser controller which I use to run Calamus 1.09n. If the laser printer is switched off but connected to the Atari my machine acts very strangely when the Atari hard disk driver is booted off my 40Mb internal hard disk. It seems to think it has additional drives attached and normally ends up bombing out or presenting a desktop with no drive icons. When I am not using my laser I have to pull the DMA cable out of the back of the Mega STe which is a pain and is not doing the connections any good. Both the Mega STe and laser work perfectly otherwise. Do ACSI devices have to be terminated like ordinary SCSI devices? I don't want to keep my laser switched on all the time. Is this problem related to the SLM605 laser? There seems to be very few of them about. I have only seen one other guy has got one and he has exactly the same problem as me. I would also like to see a general article about current hard disk models. I am going to get a 1Gb unit to replace my 40Mb Quantum drive. I bought a used IBM 1Gb drive but could not get to work despite three other members of the Wessex Atari User Group having a play around with the original Atari HD software and ICD software (thanks anyway guys). I have also been told the Quantum Fireball SCSI drives don't like Ataris although I have heard of one guy with an STe with a 9Gb drive. It would interesting to know if Ataris can use some of latest drives being released this year which are supposed to run up to 6Gb with IDE and 23Gb on SCSI! Sounds great if you want to run a BBS, burn some writeable CDs. and do some direct to disk recording all at once! ** Italics on ** John Hayward, Portsmouth ** Italics off ** ** A ** If the SLM Laser printer and controller are connected to the ASCI connector they must be powered up otherwise you will experience the symptoms you report. There used to be a hardware widget which got around this problem but I haven't seen one for a few years. The simplest solution, which minimises wasted power consumption and keeps the laser printer quiet, is to open the rear flap just enough to trip the sensor - if necessary use something to wedge it open in that position. To use the printer simply shut the flap and the printer resets itself ready for use - crude but effective! ** SLMFLAPG.JPG here ** Almost any SCSI hard drive mechanism can be persuaded to work with Atari hardware given the right combination of termination and software, which usually means using HD Driver software - refer to the SCSI Masterclass Q&A item in AC#5 for more details. IDE mechanisms can also be connected by fitting one of the TUS IDEal interface boards. ** BC ** Joe Connor ** /BC ** ** Q ** WWW & CAB 2.5 I'm running this combination on my TT fitted with Matgraph Coco Card and NVDI. I have recently experienced occasional problems trying to access links or back/forward page resulting in system locks with the error message Runtime error and what looks like a line of program code. Also when on Teletext Website error messages "Too many fonts" and on another website error message "Too many frames". Is there any way of saving HTML pages as ASCII text pages, or do I print them out then run the printout through a scanner and OCR them? ** italics ** Nick Ashby via Email ** /italics ** ** A ** The numbers following the Runtime error message relates to a Pure Pascal error code. For example, the most common error is "Runtime error #163" which means "bus error" - a crash to you and me. The problem could lie with CAB, CAB.OVL or STiNG so they're not much help to the end user. It's probably worth trying a different CAB.OVL file because there are so many different releases. You certainly don't have to print out HTML documents then OCR them to get at the ASCII! In the Options/General CAB dialog you can set up a mouse-button combination to call up a pop-up menu, which includes the option to save the source code of the HTML file. You can now load this file into Everest, or other text editor, and do a Search/Replace with leaving the Replace field blank which removes all the HTML tags leaving mostly ASCII. There are also various stand alone utilities to convert HMTL documents to ASCII which work with varying degrees of success including: ** UL ** * Egale, the latest version includes an HTML to ASCII converter. * HTML-Help, programmed by Matthias Jaap, which offers lots of options. * HTML2ASC, a no frills TOS based utility. ** /UL ** A CAB module to convert HTML source to pure ASCII would be useful. ** B ** Joe Connor/Alexander Clauss/Oliver Broadway/Chris Good ** /B ** ** Q ** SCSI dump In the Digital audio article in AC#9 Shiuming Lai suggests using ExtenDOS Pro on a Falcon to SCSI dump data from a CD-ROM to bypass the degradation which occurs traversing the analogue/digital boundary and would like confirmation this is also possible on ST/STe machines? ** italics ** Paul Keating, via the message service ** /italics ** ** A ** Yes, you can use SCSI dump on ST/STe machines, it's a simple data transfer operation which is nothing unusual in a computer system. ExtenDOS has included this feature for a number of years now, the Pro version should meet your requirements, otherwise check out the press release of the latest update, ExtenDOS Gold. If you're using the ST as a sound source (running a sample sequencer, perhaps) then you're restricted to 8-bit resolution. The benefit of dumping from CD would be offset by having to downsize the resolution as well as convert the sampling rate (if you're bothered about maintaining the speed), although it'll still outstrip sampling in 8-bit, so you might want to consider the Replay 16 sampler. This gives 16-bit monophonic I/O capability for applications with a Replay 16 option. If, however, you are using the ST simply as an editor or processor, and intend to upload the data to a MIDI sampler or other professional device (see the sample player reviews in the PD/Shareware pages), you can opt for the full digital solution, a CD-ROM drive. This will also give you convenient access to lots of software including Floppyshop's music utilities compilation. You may need to use one of the current SCSI host adapters because many of the older models don't transmit the control signals required for SCSI dump. ** italics ** Shiuming Lai ** /italics ** ** Q ** ** Leave this one out if not enough space ** SIMMs explained First of all I must congratulate you on your continued support of the Atari platform and long may it continue. I also read with great interest the Atari Computing Convention this coming November which I plan to attend. I am writing to ask a couple of questions concerning SIMMs: 1) Apart from price, is there any difference between 3 and 9 chip SIMMs (performance etc). My STe currently contains two of each and seems happy. 2) It has recently come to my attention that it is possible to buy 4Mb 30 pin SIMMs. Would it be possible to install four of these in my machine to give me 16Mb? If not, why not? ** italics ** Peter Bray, Ashby-De-La-Zouch ** /italics ** ** A ** Thanks for the encouragement, which is always welcome - see you in November! ** UL ** 1) I'm not aware of a price or performance differential between SIMMs equipped with three or nine chips, most will work fine in your STe. The only issues to be aware with concerning SIMMs are: a) Parity or non-parity SIMMs are available and both work fine in ST machines (it's an issue on other platforms). b) SIMMs have different speed ratings, most 1Mb 30 pin SIMMs will have a speed somewhere between 150ns (nanoseconds) and 70ns and these should work in any combination in ST machines (again it's only an issue on other platforms). 2) It's not possible to use 4Mb SIMMs in place of 1Mb SIMMs because the memory management unit used in the STe, although redesigned from the one in the STFM, still only had 10 address lines. The number of address lines dictates how many memory locations can be accessed, so because there are only 10 lines, only a maximum of 4Mb can be addressed. ** BC ** Joe Connor/Mark Wherry ** /BC ** --- ** boxout ** ** Vital because on Reader Disk ** ** Tip ** HP Deskjets - the missing last inch solution! ** italics ** Rod Jenkins reports... ** /italics ** In AC8 Les Charles mentioned the "age old problem with HP Deskjets not being able to print the last couple of lines at the bottom of a page". Courtesy of Wilfred Behne of Behne & Behne Systemsoftware in Germany, there is a solution - for NVDI users at least. NVDI is supplied with a utility called MAKEPRN. This program allows the user to set the margins of the paper, or so it appears, but Wilfried explained MAKEPRN does not change the margins - it describes the physical margins of your printer. What you have to do is to insert the control code which sets the printer for the page size you are using after ESC "*rbC" and the problem is fixed! To do this run MAKEPRN, select the HP 550C driver, then in the Control Characters page edit the Page start line as shown: ** MAKE_PRN.GIF here ** ** caption ** l26A sets A4 paper, other page sizes require different values. ** /caption ** ** ON READER DISK banner here ** Using this information it should be possible to adjust other editable printer drivers. For more details along with a table of codes for other page sizes plus details and tips for colour printing check out the illustrated HTML format article on the Reader Disk. ** /boxout ** --- ** boxout ** ** I've put this one a boxout to avoid breaking the NP lines... ** ** Tip ** File attachment woes? If you get an email which looks like this: ** np ** If=20you=20get=20an=20email=20which=20looks=20like=20this=20check=20the= =20header=20for=20quoted-printable=2E=20Another=20giveaway=20with=20fil= e=20attachments=20is=20odd=20line=20lengths=2E ** /np ** You've been sent an email in "quoted-printable" mode, usually denoted in the mail header: ** np ** This is a MIME-encapsulated message --defa1ca4-06b7-11d2-afce-00805fbe60fa Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable ** /np ** The same problem can afflict MIME and UUencoded email which means they will not decode correctly. Quoted-printable mode is designed to do to text files what MIME and UUencoding does to binary files, which is to ensure only 7-bit ASCII characters (32 to 127) are used. This is achieved by converting any high-ASCII characters (an umlaut, say) to the form "=XX" (without the quotemarks), where "XX" is a 2-digit hex code representing the character in the Latin 1 character set. Happily MIME and UUencoded file attachments don't contain any high ASCII characters to convert but in quoted-printable mode the equals characters "=" are changed to "=3D". This is obvious in the header of a UUencoded file if the character set is defined at the start of the encoded section: ** np ** table !"#$%&'()*+,-./0123456789:;<=3D>? @ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_ The first line should read: ** np ** !"#$%&'()*+,-./0123456789:;<=>? ** /np ** It's easy enough to use the Search/Replace option in a text editor to replace all "=3D" strings with "=", which I did, but the decoding still failed at the following line: ** np ** M=3D'7=3D@/O8P9<4X.N#;=3D=3DF)?1W7"G#UP8OKO @U.&+ZX<=3D]=3DXZ9=3DA8I-]=3D= G!]St ** /np ** Which after replacing "=3D" strings with "=" became: ** np ** M='7=@/O8P9<4X.N#;==F)?1W7"G#UP8OKO @U.&+ZX<=]=XZ9=A8I-]== G!]St ** /np ** The resulting line was still one character longer than most of the other lines which provided the clue to a further problem. Thomas Binder, the Thing author, who sent me the file, provided the answer. The other peculiarity of quoted-printable mode is that over-length lines are broken using "=" in place of a hyphen at the end of the line to indicate the break. The problem line contained so many "=" characters that when these were replaced by "=3D" strings the line became too long for the mailer and was broken, with the remainder following on the short next line. The line should have read: ** np ** M='7=@/O8P9<4X.N#;==F)?1W7"G#UP8OKO @U.&+ZX<=]=XZ9=A8I-]=G!]St ** /np ** The extra "=" at the end of the first line has been deleted and the short following line has been brought back up. After that file decoded fine! Armed with this information you should be able to solve similar problems when file attachments are sent in quoted-printable mode, to summarise: ** UL ** * Replace all "=3D" with "=" * If there are any odd length lines (62 characters in UUE format) ending with "=" and a short following line, place your cursor before the final "=" and press the [Delete] key twice * Repeat as necessary. ** /UL ** ** italics ** Peter West ** /italics ** ** /boxout ** --- ** boxout ** AC#9 Bug Report Digital audio errata The S/N figures for the STe and Falcon given last issue were only based on theoretical calculations, which (as explained in the article) don't correspond exactly to real life. Lab tests have shown factory-fitted Falcon CODECs are of wildly varying quality, averaging an output S/N of about 80dB. Please accept my personal apologies for these errors. ** italics ** Shiuming Lai ** /italics ** On page 35 in AC#9 it is mentioned CAB doesn't support secure transactions yet I've been using this feature for a while now so that must be wrong! ** italics ** Mario Becroft ** /italics ** ** /boxout **