Storm

Hello Paul! PF> Well, I didn’t like this: PF> 1. cursor brake Let's fix it. PF> 2. the screen is somehow too small some people criticize XAS for this The maximum that will happen is 40*22 (for turbo). Those who criticize XAS for _this_ are in another boat. TO_ALL: this is SPECTRUM, darlings, and not text editor, besides... If something that is not very necessary promises pro- progressive hemorrhoids, we can do this we won't. Bridges burned - storage format text _not_allowed_ lines longer 40 characters. If you try to please everyone, you will succeed bad. For example, ZXASM - to text the editor is not up to par, and the assembly inconvenient - not fish, not meat... PF> 3.hotkeys needed They will, but not at the cost of abandoning "" pr. Will be via CAPS SHIFT - transition large/ small letters are very rarely needed (there will be only via LOCK). PF> 4.marks are almost invisible It will be setup, it will be... PF> Here are the wishes: PF> 1.make it yourself 64 characters We won't, see HELP. PF> 2. if you decide to support race PF> wider memory, then try PF> support all extension options Let's think about it. The point is that when compiling to search for a tag and write down the codeyou need to click your beak. In addition, search tags actively uses the stack and uses everything registers. Therefore, CALL RAMDRV is not supported walks... And if perverted, then the time of com- hair removal will increase by 10 percent... We spared the nerves of the owners of tricky machines and did not use OUT (#FD),A, although they freed sometimes it was problematic to keep BC... TO ALL: Does anyone have any information on which machines does not pass OUT (#FD) with the installation a lazy sixth bit? There will definitely only be SCORPION support. Possibly PENTAGON 512. For other machines there will be a detailed docket, what to put where to make it work... Still, we shouldn’t do... PF> 3. The converter would not interfere. Why? See letter to Vanya. PF> Yes, by the way, even though this is a demo version PF> but still, I typed in Storm: PF> LD a,(ix+0) PF> and so assembled in all memory PF> noted the time, the same in xas, and suddenly PF> xas is 2.5 times faster, why? The way you checked, check no-no!!! a) I checked on two _real_ texts I gave my result in the help. b) we have the same amount of memory for text wow, here’s a problem - it’s getting into STORM _DOUBLE_ there are more similar lines. LD A,(IX+0) XAS: 8 bytes STORM: 4 bytes I checked for two thousand rows.Result: XAS 1.9 sec STORM 2.5 sec Indeed, slower... 1.3 times Somewhere slower, somewhere faster... Test it on other teams. I repeat, for real texts it will be 1.5 times faster, for texts with there will be more tags faster due to quick search... d) certificate. byte/line LD A,B 3 LD A,#50 4 LD A,KKK 5 LD A,KKKDD 7 LD A,(IX+5) 4 LD A,(IX+SH) 5 LD HL,#8000 5 LD HL,(#8000) 5 JR PR10 5 BIT 7,(IY+10) 5 LD HL,(#8000+ECODE)*10 14 LD HL,((#8000+ECODE)*10) 14 Question: Maybe you can tell us in general terms the methodology? compilations (you can use soap). I have: suction of one/two operands, defining the group of each operand, quick conversion according to the application table in relation to the command, then, for the first pro- stroke length is simply taken from the table, for the second one there is a transition to one of 50 code generation procedures applied to the teams. Compilation occurs in the buffer 192 bytes, overflow occurs dumping code into pages. Always open page of text. Search for a mark clicks the key vom. To search for a tag, do the following: (HASH search on SPECTRUM scale...) - all letters of the label are clogged with each other homo with shifts, we get characteristic byte, also known as index in table pointing to the first element with such a character. byte. Next, search for the typePOP HL:SBC HL,DE:JR NZ,... In the worst case (approx. 2000 marks) interruption there are less than 30 marks. Each element contains an address link to next element. The element contains everything letters except the last one (no violinist needed). The link bit is reserved, you can view Find tags that are not referenced anywhere. PF> ps. when will you do the packaging in mmd? PF> is already there? Almost... After ENLIGHT, probably... Best wishes, -=LD=- X-TRADE GROUP.

Hello Dmitry! PF>> 1. cursor brake DL> Let's fix it. By the way, it would be nice to do a survey in two minutes, i.e. first time pressing test, second confirmation .No need to repeat and then some have film keyboards double buttons are not pressed (ext,del...) PF>> 1.make it yourself 64 characters DL> We won’t, see HELP. Smart people will come along and mutilate... PF>> 2. if you decide to support race PF>> wider memory, then try PF>> support all extension options DL> Let's think about it. The fact is that during compilation DL> to search for the label and unsubscribe the code DL> you need to click your beak. In addition, search DL> tags actively uses the stack and uses everything DL> registers. Therefore, CALL RAMDRV is not under - DL> walks... And if perverted, then time com- DL> hair removal will increase by 10 percent... DL> We spared the nerves of the owners of tricky machines DL> and did not use OUT (#FD),A, although free DL> sometimes dealing with BC was problematic... Yeah, in xas because of this out (#fd),aand used and modified for more memory It will cost me a lot... DL> TO ALL: Does anyone have info on which DL> machines do not pass OUT (#FD) with installation DL> with the sixth bit? The sixth bit operates in Scorpio and on top of that, it doesn’t give anything anywhere else. DL> There will definitely only be support for SCORPION. DL> Possibly - PENTAGON 512. DL> For other machines there will be a detailed docket DL> what to insert where to make it work... DL> Still, we shouldn’t do... :) and it’s true, I saw the drivers for Alasm Kondraty (not M. Kondratyev:) almost enough. It works even worse... DL> LD A,#50 4 DL> LD A,KKK 5 DL> LD A,KKKDD 7 DL> LD A,(IX+5) 4 DL> LD A,(IX+SH) 5 DL> LD HL,#8000 5 DL> LD HL,(#8000) 5 DL>JR PR10 5 DL> BIT 7,(IY+10) 5 DL> LD HL,(#8000+ECODE)*10 14 DL> LD HL,((#8000+ECODE)*10) 14 I'll have to format the text optimize :) DL> Question: DL> Maybe you can outline the methodology DL> compilation (you can use soap). DL> I have: a hickey in one/two operands, DL> defining the group of each operand DL> quick conversion according to the application table- DL> relative to the command, then, for the first pro- DL> the length is simply taken from the table DL> for the second there is a transition to one of DL> 50 code generation procedures applied DL> to commands. Compilation occurs in the buffer [skip] a) I didn’t write the xas, I made it with version 7.2, optimizing, etc. b) while I was doing the normal conditional broadcast, almost crazy to figure it out I still don’t understand what’s what. It started like you did then there this... after the definition a specific team begins a nightmare, I'm sorry, but looking at him study before writing a letter I don’t really want him anyway. Well, you probably know about tags, but I'll write. The label table is corrupted. way page [6] divided in half, top half with #fffd there are tags starting from a to m, bottom from #dffd there are tags starting from n to z. Label format [value](3)(7)(14)byte labelsince the value is fixed, then max. the number of tags is (3)3276 (7)1820 (14)1023, macros in the same bank [6] who are killed in case of great number of tags. When compiling the first character of the label is taken and determined the required half, then brute force :( (need to be replaced) I would be glad if I could help in any way. ▌▌║▌█▐│▌▌▐▐ WiTh The BeST wIsheS . ▌▌║▌█▐│▌▌▐▐ ▌812/03.00▐

Hello Paul! PF> You don’t have to do it on repeat PF> and some have film keyboards PF> double buttons are not pressed. What do you mean? The buttons don't work at all and do you have to click through the main ones? So this is the problem of broken keyboards... PF>>> 1.make it yourself 64 characters DL>> We won’t, see HELP. PF> There will be smart people and they will mutilate ... The process of mutilation will serve as punishment for them. I eat... It’s been said that the devil will break his leg... DL>> PO ALL: Does anyone have info on which DL>> machines do not pass OUT (#FD) with installation DL>> with the sixth bit? PF> The sixth bit acts on the current in Scorpio PF> and on kai, it doesn’t give anything anywhere else. This is understandable. But, say, on PENTAGON 512 sixth/seventh bits are addressed additional pages? So if you do OUT (C), #10 first, and then volume OUT (#FD),#50, then will be addressed different pages. If I'm wrong maybe who will explain? And about other cars too... PF> I'll have to format the text PF> optimize :) Hemorrhoids are a man's friend, comrade and brother :) PF> Well, you probably know about labels, but PF> I will write . [...] Thank you. But, I think, not everything is so simple, a simple search takes a very long time. In ZXASM and as far as I know, full alphabetical sorting with link to the next element. Probably, XAS has a very clever overkill... Best wishes, -=LD=- X-TRADE GROUP.

Hi, Alex! AM> I hope someone has seen this assembler from ZF-6?;) AM> So, there is a very small glitch: if it is present in the text AM> such a combination as, for example: DB "1", then during assembly AM> all labels are shifted one address up, i.e. if it was: Yes, there is such a party. We'll fix it in the next version. Here is a list of detected glitches: DB "A" LD ($), SP; exactly ($), it's terribly buggy (1<2)*1792; the expression is considered incorrect "FREE" is misspelled when the value is from 10000 to 11000 The rest is smaller and insignificant. By the way, I heard that someone has STORM The discs rattle steadily. So this happened with the old version (32 characters per line). It seems that this is not the case in the ZF version. If anyone notices anything, write to us and we’ll fix it. Best wishes. Dmitry aka -=LD=- / X-TRADE GROUP. -+- ZXASM 3.0

Hello, hello Leonid! One day, in the chilly summer, something around (08/22/98/04:57:08) Leonid Mishankov once wrote to Vitaly Vidmirov. VV>> I wish I wouldn’t use it, because ~30% of the code is _mine_!!! LM> In this case, can you tell me which of you wrote the disk driver for it? LM> ;-) What, are you going to hit me? 8) LM> (he got me - the slightest glitch on the disk - and neither read nor write :( How is the disk _driver_ responsible for the presence of glitches? on your disks? His job is to read/write sectors without errors, which, by the way, he copes with successfully... i.e. Storm catalogs does not spoil and registers sectors normally, although it is possible that there will be on fecal drives... By the way, Storm is not a disk doctor and cannot work with broken disks obliged... By the way, I haven’t experienced any glitches due to software reasons. for several years - since when I erased all copies conver comandera from disks and switched to FPM (greets2MaxPetrov!) And since then, when I installed a normal disk drive - I also stopped having hardware glitches. reasons (although I had to switch to Direct comander/DR!). I wish the same for you... Evil Vitalik AKA Dark / X-Trade