trodas

Also to submit the score, it require "This benchmark requires a data file. Please upload the saved benchmark result." BUT there is no clear button in the HWbot edition of Aquamark to save the benchmark result! The menu also did not show anything usefull at all. "Save to desktop" button, hidding partly bellow the GPU-Z did not work at all (eg. NO file is created): ...so I tried out of desperation (installing new stupid WinXP, tweaking a little and this THIS!) the Debug and save the log file, but that file is rejected by HWbot website and also containing NO info about the benchmark... Included. PS. there si absolutely NO file in the aquamark directory logs after the bench run too. Please advise how to do it... PS2. found in Documents dir the Aquamark3 directory and some files in it (Config.txt, License.txt, Log.txt and Run.txt) and also a directory "Benchmark 2013-12-17 13-30-43" - yet NONE of the files inside that directory is accepted by HWbot www page...! Including...

To be able submit the Aquamark score (pretty weird, since original Aquamark run well on my Win2k SP4) I had to replace Windows on my testing rig to Windows XP SP3. Now from that point on, I was under the firm impression, that everything will go smoothly. I was... wrong again. I downloaded the latest HWbot Aquamark (v2.53.108) edition install and installed it. With all the options (yes, including these libraries, that SHOUDL be included by default IMHO with the app, so install is not need...). Started it and got a message that the application cannot run, because it need MSVCP100.dll ...! And that DLL is not found. Searched whole C and D partitions and it is right! The DLL is just not there... Tried the M$ Visual C++ libs install 2010, but no go Weird. So I ended up downloading the need MSVCP100.dll and MSVCR100.dll from DLL files - check the latest versions under the "alternative files": http://www.dll-files.com/dllindex/dll-files.shtml?msvcp100 http://www.dll-files.com/dllindex/dll-files.shtml?msvcr100 ...and then Aquamark finally run. The machine info is almost non functional, so if THIS is the added benefit of HWbot edition, then we will be better w/o it... Sorry to say, that was just my experience DLLs to make the Aquamark work included. Somebody should look at this, when it does not work fresh WinXP SP3 install (yes, I disabled some services and WFP, but that it is, no more mods (yet) ), then it is IMHO disfunctional. Once again, old Aquamark worked just fine on Win2k SP4. I see no reason why it should require the slow WinXP... (and yes, WinXP SP3 is _TAD_ slower on Duron CPU by just moving the windows around, not mentioning the rest of the stuff)

Antinomy - screwed L1 mod - that was my best bet also. So in order to test it, I removed the wire mod that took me so hard to get it right and begin testing. So far, every multiplier I tested (5, 5.5, 6, 6.5) worked w/o a glitch, when set on the jumpers on the JetWay V266B. How many more I have to test so we are sure that L1 mod is fine? In the process I managed many interesting results, like 155MHz FSB for Duron: http://valid.canardpc.com/jqxizm 974MHz out of the 750MHz Duron: http://valid.canardpc.com/1uilhv ...and maximum bench stable 148x6.5 - 961MHz: http://valid.canardpc.com/ziaske Failed to get the SuperPi 1M under 2min, tough. 4 sec shy... In DOS, it can run Memtest at 988Mhz (152MHz FSB x 6.5) ... Regardless it looks like the multiplier changing does work as intented. The only problem is, when I set the multipliers to AUTO - on the JetWay V266B. On the target MSI 6340 it is set by this by default forever and nothing one can do about it. Then the board (like the MSI 6340) fail to post! Therefore I'm affraid that I need to do something else - like force some default multiplier or something like that. How else to test the L1 mod? I got in the mind the JetWay V600DAP - it works well at 200MHz FSB, witch is far more the Duron can handle it (x5 at 200MHz give 1000MHz and the Duron just failed SuperPi 32M at 961MHz, when it did fine at 1M test at the same clock just minutes before... probably need better cooling for such clocks and voltages) - so I can push the maximum Duron FSB there for sure, it will top because of the CPU unstability at about 190, 192MHz... Yep, I have there the AXP 2500+ with the cuts between then L5 bridges. I apply some superglue to the cuts to fill then and then try connect the two bridges to have the mobile CPU with multiprocesor support out of this AXP and then we can test out what happens on the MS 6340 and the JetWay V266B multiplier settings. I'm worried that the wire setting does not work, because it was always in my mind that I wanted to try convince the AXP in JetWay V266B to run at x14 multiplier and the mobo top at the x12.5 multi. So by leaving the jumpers at auto and forcing the wires I hoped I can run the AXP 2800+ at it's intented 2100MHz speed (150x14 = 2100) ... realizing that there are problems make my dream much harder to achieve now. IIRC there are some startup settings and then these bios settings (equal to mobo setting?) and even flipping the low-high multipliers too... sure there must be something wrong... Turrican - not to worry, I get the joke. I'm long time not young by any means and I was not modding enought around when the AXP come by, so I making it up. And stumbling upon a hardware, that give around 33MHz (or 35? dunno) FSB give me plenty of ideas what to use that for Just get the low multiplier

Well, I did not like "just a pencil", I wanted a permament, good solution. Therefore I used the conductive paint to get a good result. On the top of this, I going to apply a "drop" of normal, transparent paint to protect the conductive paint from clearing, so... as far as L1 bridges connection goes - it does allow me to choose any multiplier I want in the JetWay V266B (and possibly in other mainboards) - witch IMHO proves, that it work well. Nope? However the catch is, that WHEN the multiplier settings are on the mainboard set to AUTO, then it fail to work! And the AUTO settings is exactly what IS recommended on the OCinside page and also the only one settings the MSI 6340 can have... So I hit a wall there Did I use right settings for this x5 multi wires? http://www.ocinside.de/html/workshop/pinmod/amd_pinmod.html

I did connect all the L1 bridges, that was the start of the experiment, mate. I write that in the first post: The solution looks obvious: connect all the four L1 bridges to get multiplier unlocked Duron. So I did that today with conductive paint by this guide The mainboard choice is also explained there. Can Abit NF7-S/AN7 do 33MHz FSB? As the original problem, I vaguely seems to remember something about starting multiplier or something like that... that could be the culprit there. Help is need, tough...

Thank you! Time to add some interesting scores, lol Who says I overclocking? The final clock is still 750MHz: http://valid.canardpc.com/h0ivt2

HWbot is missing yet another mobo - JetWay V266B. I recapped the mobo long time ago ( http://trodas.wz.cz/index.php?act=ST&f=16&t=355 ), got another one and testing now some things with multipliers, and ... it should be added It use crappy VIA KT266A chipset, but at least it can manage 150MHz FSB with 12.5 multiplier, so it is not that bad. However JetWay did not want to know about this their mainboard, but I do still have the manual in pdf form! A proof that I did not make this mobo up, lol: http://www.mediafire.com/?in41jn2dzt6kimi (JetWay V266B user guide manual) There is best picture of original V266B I managed to find: (mine is now w/o many ports and with nonstandard caps and even few polymers near AGP, so...) Other pictures not usable for HWbot: http://cdn4.asteroid.pl/c17/a.swistak.pl/217/219/217219356_1024.jpg http://allegro.logosms.webd.pl/fotki_01/plyta-glowna-jetway-v266b-via-kt266a-3-jpg-Allegro130411.jpg (notice different caps, mine got all GSC supercrap caps...)

So I'm confused, why it does not want to work. I tried this - since the MSI 6340 is obviously no-go, then I look at my wiring job and find out that there are few bad wires making contact where they should not (used too much wire, mostly, even is the size of hair, there is still too much for 4 threads around these pins, I suggest 2 max and end in middle of these), so I fixed this and tried again. Again no go. So I get worried if I did not blow up something and did another test - back to the JetWay V266B, that worked. Now it does not. No way to post... I was clueless, but then I tried something that is quite interesting - instead of having the multipliers set to auto, I tried set them directly to x5 multi, as I believe I had already set in the wires bellow the CPU. SURPRISE, it posted and even worked fine at multi x5 ...! WTF!!! So I tried x5.5 - worked again, but at x5 settings. Then I tried x6 or even auto again, but none works. So as long, as I set by manual anything x5 multi (x5, x5.5) it work fine in the JetWay V266B mobo. Hmmm. That is NOT what I was hoping for when I connected the L1 bridges! I hoped I can set any multi I want and force it by wires! Obviously this is not true... Could someone tell me why? And one more interesting thing - on post, the Duron 750 is reporting itself as Duron 1400. No kidding. x5 or x5.5 settings, always Duron 1400 (I never knew I could upgrade Duron 750 to Duron 1400 just by few wires... but it work as 500MHz, so, slow, lol) It just make me think, what is MSI 6340 try to run it at 1400MHz settings? Eg. it is obvious, that the mainboards probably go for the fastest multi (when multipliers are free) and ignore my wires. What should I do?! I hope not what I think - eg. change the L4 and L6 bridges... I did not exactly want permamently x5 multi. That is why (damn!) I was connecting the L1 brigdes, so I can choose what multiplier I want. And upon adding the wires is worked (on mobo, that allow the multipier changes, that it is) ... Help, anyone? As far, as the x5 multi goes, well, there it is: CPU-Z Validator 4.0

Guys, since I discover a bug in mainboard, that crash down the FSB to 33MHz, then it making that mobo ideal to do some very slow runs Now the weakest CPU I managed to source is a Duron 750. That is 100 x 7.5. When it is 33 x 7.5 ( 248 - 262MHz ), then it is still too "fast" (SuperPi 32M 8h). http://valid.canardpc.com/evdcpe So I looked at how can I lower the "high" x7.5 multiplier. The solution looks obvious: connect all the four L1 bridges to get multiplier unlocked Duron. So I did that today with conductive paint by this guide: http://www.ocinside.de/go_d.html?http://translate.googleusercontent.com/translate_c?depth=1&hl=en&rurl=translate.google.com&sl=de&tl=en&u=http://www.ocinside.de/html/workshop/socketa/socketa_resistors_d.html&usg=ALkJrhgFicC6QnLDdYiJzx32ObTgfnt_Dw I was sucesfull I think, because on JetWay V266B mainboard I managed it to work well. (there are 5 switches that allow me to change the multiplier from x5 to x12.5) running at default, at x5, at x5.5 (with FSB 133) and x6... (x8 the CPU did not make, quite poor overclocker (!) but whatever) Then I moved to the target board (MSI 6340) and after inserting the CPU, there is nothing new in bios - eg. no multiplier select table, nothing like that at all. Suxx! Yet I was ready for this eventuality and with the unlocked multi I wanted to force the multiplier down to x5 by wires from the bottom side of the CPU around the pins, by this guide: http://www.ocinside.de/go_d.html?http://translate.googleusercontent.com/translate_c?depth=1&hl=en&rurl=translate.google.com&sl=de&tl=en&u=http://www.ocinside.de/html/workshop/pinmod/amd_pinmod_d.html&usg=ALkJrhgfaKJz4biu1Us66eSygvAftUhGWg And that is, where I hit the wall. The MSI 6340 refuse to post and it make me wonder, if the multipliers are any different between the Duron and the awailable settings for CPU's - AMD T-Bred, AMD Barton and AMD Athlon XP. I think not (there is no way or reason why they should be any different and even the interactive pin guide show no sign of and change between the CPU's as far, as the multipliers go) but the MSI 6340 is not posting now. Not a beep, just fans on, but that it is. Anyone have clue about what is going on? I go to check the wires under the CPU and if they are fine... then perhaps test with the JetWay V266B, if there the enforced x5 multi will work... Ideas welcomed. Or the KT133 cannot work with x5 multiplier?

Christian Ney - Oh, that is a sad look. A new PSU 24pin connector you can get from Mouser.com, just use search for this item number: 571-215860374 (PSU AXT 12V 8pin got this number: 571-1-794305-1 ) Good luck replacing it and fixing the cause. Mining should not be as power hungry, that it fry the connector... yet I read specs of these connectors and oddly enough, they say only 6A per pin, so... I quess we kinda overloading these in our computers

Oh... january and only start... well, we have to wait, sadly. Let's wait for january then - it is not much days to wait and the graphic and editing quirks seems to be easily fixable - I hope.

Oh, well... hopefully it won't took till next Xmass to become fixed DDR3 bug also seems to insist on staying as well, as the inability to edit overclock changed in GFX card when using the prepopulated submissions to save time... Do this get at least higher priority?

Thank you very much! Now is time to provide some results P.S. Looks like the image problem is still far from being fixed... any news on this expectable for Xmass?

It is not THAT bad board - can do 2300MHz AXP: http://valid.canardpc.com/t87qtk ...witch translate to 47sec SuperPi 1M test... :*)

Suprisingly, HWbot: Your search results for "JetWay V600DAP": - none! Interesting. I recapped the mobo long time ago with pre-production Samxon caps and they turned out to be bad... Yesterday I managed to fix my JetWay V600DAP mobo to have one stable working testing AGP mainboard - I used for ram/AGP part testing. It was just a test, so, no wonder. Yet now I need it and it did not even posted with my PNY 6800GT, so... after the recap it post and work just great, even with this graphic card, so I'm happy and going to work on it for some tests too AMD Mobile CPU can run in this old mobo at 2300MHz stable w/o problem at 1.575V I wish I had a better cooler, it might go higher Oh, HWbot don't know this mobo! So, pls, add it. Even JetWay did still know about it: http://www.jetway.com.tw/jw/motherboard_view.asp?productid=29&proname=V600DAP And there is best picture I managed to find: (mine is now w/o many ports and with many polymers = not original anymore, lol)

8GHz on AMD - ownage Under 10min on SuperPi 32M test? Very respectable. No wonder... (besides, flanker is flanking this result of his friends all around the AMD crowded forums, so...)

Bitcoin: What You're Not Being Told Bitcoin major design flaw, that is inherent to the system, is the way it protect itself from used being able to use own Bitcoin twice. The way it do is to maintain blockchain, witch is list of all transactions with every Bitcoin since beginning to this second. And that list for every Bitcoin users on the world! Add there security hashes and you get exponencial grown of size of this blockchain. In August 2012 is was 2GBy. In August 2013 it was 9GBy. ...and right now we are at 11,9GBy. As anyone can see, with such exponencial grow of the size (the more Bitcoin is used, the more the blockchain grow) it this unsustainable: Bitcoin Blockchain Size For example on Bitcoin web they do admit, that it will be necessary to use 14TBy (!) space on your HDD for each 85 days of using Bitcoins, it they are used on the same scale as Visa cards: https://en.bitcoin.it/wiki/Talk:Scalability#Disk_space Now everyone who wants to use Bitcoin, it must todays download to own harddrive whole 11.9GBy blockchain and the size of the blockchain grow exponentialy and soon it can reach hundreds of gigabites and then terabites, witch means that for average user it will become useless. And when this happens, then whole good Bitcoin idea get flushed down the toilet, together with the wealth some people are investing into Bitcoins. Even that I like Bitcoin very much, it is obvious, that this major design flow must be fixed and removed. Yet on the solution was not reached agreement even between the developers, let alone the fix is getting implemente. And because possible solution with "supernode" (sometling like a bank for Bitcoins) open Bitcoin to the attacked and later controled by governments, because such "supernode" will not be anything you can run in your garage. The harddisks requirments will be nothing except really extreme. For example the promised (not materialized) 60GBy drives it will eat like crazy. If it will be used like Visa, then it is need 14TBy of HDD space for each 85 days, so that translate to 239 pieces of 60GBy HDDs. To put this into perfpective, it will require that 2,8 of these HDDs will be added to the system each DAY! This is not sustainable by any measures. Just the harddive failures nail end to such "supernode"... and then Bitcoins at all. Imagine then that there is a catastrophy (gov sieze the "bank", fire, flood... sabotage) and then everyone will lose everything they have in Bitcoins. And all this does not take into account that these 60GBy drives are not there and we have no idea how reliable they are... Security expert Dan Kaminsky's analysis of the scalability problem: Bitcoin: What You're Not Being Told | StormCloudsGathering I posting this a warning to these people: litecoin mining 40pcs R9 290 ...because it is irracional to see that everyone is overlooking the serious design flaw since the start of Bitcoin.

Oh, that. Then yes, the Radeon R9100 is not that slow when compared to others, even in PCI mode (and slow P4) I was not the last on the table w/o even any overclocking, so yes. That makes sense even in HWbot scores. I mean the overal todays usability of R9100 is doubtfull. For 2D it is fast enought, but that it is. There is no way that it can be used to play any more recent/decent games, that was my point. And you are right, the MSI PM8M3-V board is nightmare from start to end. Yesterday it again crashed on me when loading a level, so definitively the board is up to the blame, not the graphic card. Therefore next step - replace more mosfets in hope that this will fix it, somehow. The idea is, that during combined load (HDD, ram, GFX, sound) the powering of the northbridge get so weak to the point of crashing - and I cannot come up with anything better ATM. Sadly I lack money to get anything more usable ATM, even the caps was bought by friend in Digikey, so... No way to escape the MSI "monster" ... As for the R9100 - I will measure where the GPU voltage is and add caps there in hope for more MHz and we see

Nop, sorry, I cannot do that hard repairs. Caps are easy... Exchanging mosfets is possible only because I can heat up each leg separately, and push them up. When both of them are hanging in air, then I go work on the top of the mosfet (drain - heatsink connection) and then it move around the mobo and off he go. Clean this up, solder legs and then drain of the new one and - if done corectly, this works. But that is mosfet that have two legs and drain. In case of DrMos, you need to have all 8 legs heated up at once. I described how that can be done, but I never did that - ever. I would *LOVE* to try it, but first on dead board, where no harm can be done. This board still works... It might change, after learning how to replace DrMos mosfets Besides, I have no AM2 or AM3 cpu, so this is not board for me. Also none of us know, how hard will be to source the replacement DrMoc mosfets... Gotta be carefull with these voltmods.

Then the mobo does not (at least not with current bios) support AM3 cpu... or the support is damaged. The Vmod probably then killed just one of the phases, and the rest can do it, if you did not overclock much. Before trying to repair it, make sure you have at hand a new mobo.

About four years ago I was confronted with this unfortunate fact, that computer witch I build for friend, started giving him troubles. The image on screen started going off, PC refuse to start, etc. After a while the PC get started in the end, but it was pretty clear, that once it will fail completely and it will be impossible to start it anymore. Not even with convincing the PC to start anymore. This problem was the more embarassing for me by the fast, the my friend absolutely did no play games at all, so he is not stressing the gfx card much at all. That is why I picked up reasonably low powered (24W TDP) 7300GT with masive passive cooler for quiet operation. That is because in the case, there is a bit 120mm front fan, that at 7V spin pretty well and move cool air directly on the heatsink, right to the fins well orientated for it. So I expected a long life, despite the questionable caps (any cap, that is no genuine Rubycon, Nichicon, Samxon, Panasonic or Sanyo I call questionable if not outright bad... and yes, I did not like Chemicons and not consider them as good, but I could be mistaken or smeared away by bad personal experience with them, or more precisely - with likely fakes of them). However the cause was embarassing for me the more, that I did the PC build from start to finish and has virtually only a little budget restrictions, so I was proud of the result and... this was just unexpected. Not to mention that the GFX card itself looks cool: After we discovered, that the waiting for RMA will took a month (and my friend was not cool with this and he fear, that after a month it will get his old card back because it works and that will be it), so we took a round to local shops in Prerov, yet not find anything that will be at least closely as usable and good as this card. So I get the idea, that I could try recap the card and see, if that helps. I suggested this, even none of the caps on the card are bulging or leaking at all. Yet they are still qestionable caps and there aren't that much of them on the card, so we decided to give it a try. Rush was really big, so I did not even write down what caps are there in the first place, not even taking a picture of them. These pictures I made just after four years... As you can see, I used a combination of what I had on hand. The bigger caps I replaced with Samxon GC caps, on the smaller ones I used Panasonic FM caps. I did not even added the missing cap. Also I did not optimized the voltages of the caps - at normal situation I measure the voltages that are really in action there and then use as close voltage and as big caps, as they fit there. Trying to make the voltage filtering better. But as I was saying, there is not time. It was like a blitzkrieg. Yet despite the rush, the recap went fine and th card kicked on. And mainly - from this time on, the card is not giving any problems at all. I get to it after four years just because Mirek lend me it for testing, what works in my Sapphire A9RX480 mainboard. Because MSI GeForce 210 card do not... So after four years of usage are there no signs of any problems, that with the questionable caps come just after one year and they are quite serious problems... And the graphic card is reasonably hot in run. Temperature just on the testing desk w/o any fan around is very high indeed: So after years I checked up on my work and come to conclusion, that it was not that bad, after all. So if anyone is recapping a Poing of View 7300GT gfx card, then he need something like that: 5x 1500uF 6.3V Samxon GC d10 4x 470uF 16V Samxon GC d6.3 2x 120uF 16V Panasonic FM d5 ...or adequate replacements, like Nichicon HN/HM/HZ or polymer caps (Nichicon LE, FP...), where it will be necessary measure the actual voltages on the caps, because polymers are for 16V huge and hard to fit there. And I have my doubts that the 470uF 16V caps are under 12V. There is almost no way they will be. Usually the input voltages are used as low, as possible, to get as few voltage drop on the regulator, as possible, while this way is used to get as low heat output out of the regulators, as possible. For example the two 1500uF caps on the back of the card will be for rams, so they are likely only up to 2.5V. And the 470uF 16V caps before them will be on the input of voltage regulator, likely powering from 3.3V or 5V rail. Hardly a point for 16V cap, IMHO... etc. It can be optimized a lot

It would be quite hard to desolder and solder back these DrMos chips, as they have "too much" legs, and all need to be heated up at once. Usually such attempts end in tragedy, as the PCB give up and you tear the trace from it... But good luck on the attempt! (I would first practice on another, dead mainboard... and obtain the need parts.) IMHO the best bet would be to desolder heat-sensitive components (the 8pin ATX power connector, the top-most 100uF 16V SEPC cap on the picture), remove components that stay in the way (the two top R80 coils), shield rest with aluminium foil and use heat gun to remove the DrMos mosfets. Then solder replacements. Then rest of the components. Then cross fingers and try it... In many such cases even the control chip is damaged and there is no way to bring the mobo back... I won't not dare to do such work, because the risc is enermous. Sorry.

Antinomy - well, this project was more about getting a 100% reliable GFX card to determine, how to fix the MSI PM8M3-V mobo that is failing in the AGP / gaming only. Therefore I picked up the card that has the less draw that I get. There are a FX 5600XT too (well overclockable from 235/200 to 375/267), but it will require considerably mode power, so my main machine will become unstable... Therefore yes, the efforts to get back into working state such card is worthless, but since I did not have any 100 reliable AGP card except this one now, then I need these results (and with overclocking - even stable, tested for hour on ATItool - the board crash in gaming, so I did not fixed it yet... time to look out for another mosfet(s) to exchange to get the stability... ) You are right. But I did not attempting to get any high scores with the card. It is too slow for that to begin with However I have saved previous max. stable O/C results and they are: 250/200 - to 265/248 Now it is: 250/200 - to 290/248 That means that the rams O/C are not affected at all by the quality of voltage filtering (eg. caps), yet the GPU overclock is affected and when new, the card was actually slower that todays, after the recap Might be fun to add more caps and see, if the GPU clock can reach 300MHz or not But sure as hell I need to fix the MSI mobo problem first, because lack of good power on the AGP slot is crucial Turrican - sure he have the point - the card is too slow to begin with, but as AGP testing card it will serve it's purpose well Not much else is expected anyway. Except 300MHz GPU clock, lol.

Yep, let's hope so. Sadly poor ISON did not make it. After a while it looked like it can... but then it faded and probably get destroyed entierly. Sad. Same vay Sun killed comet Elenin: And bunch of others So it is nothing new...

So, since my machine is not crashing anymore during gaming, then I decided to try look out, what overclocking is possible with this recapped Sapphire Radeon R9100 card. And it is not that bad: It can do 293 for the GPU, but I backed it down to around 290, same for the rams, around 248MHz it can do. That make it clear, why these rams had to run at 200MHz, because they cannot take the normal 250MHz reference design clock, not with any caps. Hence I cannot match the XT version (300/300 vs 290/248), but at lest some overclocking is possible. It would be interesting to know, if adding more caps on the empty spaces (there is space for DAMN much caps!) could help to reach faster speeds. I was kinda hoped for the 300MHz mark, but I got burned shy 7MHz of it...