November 6

I would need to check a list of the technical things that the http does in the background from the point I write, for example, www.google.com in the bar and press enter, to the point when the page is loaded and visible in the browser. I know that it's a bit beyond the scope of wikipedia articles, so perhaps it should be available somewhere else. Cambalachero (talk) 02:04, 6 November 2012 (UTC)[reply]

How about starting at the Web Architecture page from the

Mozilla Firefox, far beyond the network layer of transactional HTTP. Compare the behavior of an tool like cURL, which also implements the HTTP protocol, and whose source is also available for inspection. Nimur (talk) 04:42, 6 November 2012 (UTC)[reply

]

If you mean you want to see all the file loads, scripts, etc. that happen in a modern browser when you visit a site, I suggest installing Firebug into Firefox, then activating its "network" tab, then visiting google or whatever other site interests you. The network screen shows you all the downloads and their latencies, and the inspection tabs let you examine all the scripts and so forth. 67.119.3.105 (talk) 23:21, 6 November 2012 (UTC)[reply]

OK, so Apple and Microsoft entered into a cross licensing agreement. Does this mean in 1997, and this year as well. Does that mean Microsoft can give the license to its OEMs so they can design laptops that look similar to the MacBooks? --Sp33dyphil ©^hat_ontributions 03:52, 6 November 2012 (UTC)[reply]

The exact details of the current agreements aren't known, but the 1997 agreement did include some design patents, though there are "anti-cloning" provisions to avoid making products that look too similar or have too much overlapping "look and feel". It does not allow Microsoft to make "substantially identical" products as Apple. This article sums up this part of the agreement; you can read the original agreement here if you know how to parse legalese. I very much doubt that Microsoft can farm out any of such licensing to its OEMs (and it obviously can't let them do what it itself can't do). --Mr.98 (talk) 13:20, 6 November 2012 (UTC)[reply]

Then how come Samsung and HP have been making very similar products to the MacBook without getting sued? The HP Envy comes into mind. --Sp33dyphil ©^hat_ontributions 13:37, 9 November 2012 (UTC)[reply]

I was wondering why hardly any 3G tablets support phone calls and SMS messages? Apparently Samsung's original 7" Galaxy Tab did, but no recent tablets support this feature.

I know that tablets are inconveniently large to use as a phone, but searching around the internet there do seem to be quite a lot of people who want that feature - and people already use them for Skype and VoIP apps. It would seem to be almost free for the manufacturers to add it to tablets that already have a 3G chip, a microphone and speakers, especially as the phone dialer / SMS apps are already built into Android and iOS.

But, obviously they don't, so there must be some reason. Is it somehow related to regulations, with tablets not needing to go through the same certification process as phones? Am I wrong about it just being a small software change, and a voice-compatible 3G modem is actually more complicated than just a 3G data chip? Or are they just trying to make more money from selling people two devices instead of just one?

59.108.42.46 (talk) 04:02, 6 November 2012 (UTC)[reply]

Actually a number of Samsung tablets do support this. It does seem to be a feature commonly absent. Nil Einne (talk) 12:20, 6 November 2012 (UTC)[reply]

It just adds complexity, cost, testing requirements, etc., for a feature not many people would really want (who would want that that doesn't already have a cell phone ?). Just because you can do something doesn't mean that you should. A cell phone can fit in a pocket, and hence go just about anywhere. Is the same true of a tablet ? StuRat (talk) 22:38, 6 November 2012 (UTC)[reply]

Does Technerd.com rank well enough to be included in Wikipedia? Plz consider Technerd.com`s inclusion into Wikipedia. TX — Preceding unsigned comment added by 67.49.41.230 (talk) 05:06, 6 November 2012 (UTC)[reply]

Inclusion in Wikipedia is based on Wikipedia's notability policy. Please look at

WP:WEBCRIT and then let us know under which grounds you think this website qualifies. Marnanel (talk) 19:42, 6 November 2012 (UTC)[reply

]

And Dell has one that turns the screen within the frame first. (Sounds fragile though.)

And yet, I have heard of no

But I can't know everything in the world. What models do you know of that are 14-16 inches, can handle intensive games well, can turn into tablets, and are customizable? That is, customizable enough that I would have a wide latitude of specs to configure, like in this example.

Thanks. I'm to get a smaller, replacement laptop in January and hoping it's that multifunctional. --

I'm aware of none. It's also likely that the growing tablet market will marginalize traditional convertible laptops, despite the use cases being fairly different. That said, you can get a ThinkPad with at least a Core i7, and while that might not be two nvidia or ati graphics cards bound together, it's certainly plenty. ¦

Reisio (talk) 16:37, 6 November 2012 (UTC)[reply

]

From a gaming POV, you're likely better off (both from a performance and drive compatibility POV)_ getting something with a

AMD Fusion#"Trinity" (32nm) unless you're playing a game with very high CPU requirements. (May not be the best idea from a battery life POV depending on your usage requirements.) Nil Einne (talk) 08:09, 7 November 2012 (UTC)[reply

]

BTW, I think the bigger problem for the OP is that the discrete graphics market for laptops is dying with the modern SoC CPUs with integrated graphics now generally powerful enough to play even high end games, at least with low settings (and perhaps resolution). For something like a convertable tablet which probably isn't of that great interest to those still interested in discrete graphics on laptops, you're going to find it even more difficult. And the OPs screen size requirements seem to be at the real upper end of the norm in the convertable tablet market. From a quick search, from what I can find there is only one or two current mainstream options for a convertable tablet with discrete graphics. There is the older Fujitsu T901 [1] but it only has a Nvidia 4200M which from what I've read is probably weaker then a HD 4000 let alone something on the AMD Trinity line. (There was an even older HP TM2.) There were claims during unveiling that the Transformer Book would have a discrete graphics option [2] [3] [4] and also a 14 inch but I can't find any mention of either on the Asus site [5] although may be it's still coming or the Asus site just doesn't mention them. Plenty of convertable tablets do come with Ivy Bridge I5s and I7s and some come with Trinities so your best bet is to look in to that. Nil Einne (talk) 08:42, 7 November 2012 (UTC)[reply]

I wanted to see if increasing processor frequency would give a noticable improvement (System is GA-MA790XT-UD4PF4, Athlon II X4 620, 2600MHz, HT 2000MHz, DDR 1333MHz, Ubuntu 11.10), wasn't expecting much to be honest. Since the max multiplier is locked , I changed fsb clock from 200MHz to 250MHz and adjusted the multipliers so HT, NorthBridge and memory clocks would stay the same.
Testing video conversion with Avidemux showed such a big difference, much more than the increase in CPU speed could explain, that I did another test with CPU speed as close as possible to the original value of 2600MHz; a multiplier of 10.5 gave 2625MHz, only one percent faster.
Result: 89sec to convert the test video, compared to 121sec with original settings (tests repeated several times).
Comparing the results, it seems too good to be true: CPU, HyperTransport, NorthBridge and memory frequency unchanged, only the FSB clock is increased with factor 1.25, yet I get a performance gain of 1.35? How is this possible? (input and output files were less than 100MB, I can't see I/O speed making such a difference...) Ssscienccce (talk) 08:37, 6 November 2012 (UTC)[reply]

AMD processors don't have a FSB and haven't for a long time, the clock you're setting is the base or reference clock not the FSB. If you'd adjusted all multipliers and settings to be the same, you really should not be seeing much of any difference. Are you sure you didn't neglect to take something in to account? In particular, the way the motherboard will set the memory timings when you are overclocking in this way even if you've adjusted the multiplier is not always easy to predict. (I'm presuming you double checked with software when the computer was running to ensure the clocks are what you expected.) Did you actual make sure the timings were the same (or set all of them manually)? There is also the possibility you neglected to fix the PCI express clock although I would expect instability with such a higher overclock and unless you've put your video card in a 4x slot or something I wouldn't expect it to make a significant difference. (I'm presuming when you said you repeated the tests multiple times this means you did restart at least once for the slower speed, or preferably repeat the tests after getting the higher speed ti ensure something didn't happen which screwed up the results.) Nil Einne (talk) 12:14, 6 November 2012 (UTC)[reply]

Oh I forgot to mention you may also want to make sure power saving settings aren't screwing up the result (in theory they shouldn't unless you have very aggressive settings for some reason, but stuff can always go wrong so it pays to take it in to account probably by disabling them). It get's even more complicated if your CPU has some sort of turbo mode but from what I can tell, it does not. Nil Einne (talk) 13:14, 6 November 2012 (UTC)[reply]

I recently cleaned the (considerable amount of) dust from the heatsinks and fan because the cpu overheated when running avidemux for a few minutes, so I assume it runs at full speed. I ran the test without any monitoring software straight after rebooting to make sure conditions were as identical as possible. Might check some benchmarks, the few I ran didn't seem to show any difference, but these are CPU only I think (blowfish, cryptohash, fibonacci, FFT). In any case, it's the speed of apps I use that matters.

I know it's not an actual fsb but many sites seem to call it that. My bios calls it CPU clock, same name it gives to the actual CPU frequency. Running on 250MHz and CPU 2750 now, no problems but I'm slightly worried whether the "base" frequency drives something I'm not aware of. Did find out that 350MHz is too high, had to clear the cmos after that because the sytem went dead. Something in the cpu or chipset that couldn't handle it, or the BIOS allowing values the motherboard can't generate?

It also occured to met that with four cores available, the percentage of cpu load by avidemux could depend a lot on how multithreading is implemented in the code; I have no idea how this is done in practice, but could it be that for example different code is used depending on the number of parallel threads that can be started? Say you have a "main" loop that spawns "subthreads" based on the free cores available, and small timing differences result in a subthread lasting just a bit longer than the loop so one less subthread will be started every time? Maybe I should look at the average or total cpu load avidemux uses. Will have to wait till I get an extra HD to backup my system, I want to test the 64 bit version to see if that runs better (maybe add some memory as well). Ssscienccce (talk) 16:34, 7 November 2012 (UTC)[reply]

I'd be grateful for some step by step advice on the best and least fussy way for someone who is not a techie to do this using Firefox, Chrome or IE. Preferably, the end result would be a single file in the folder. Thanks. --Dweller (talk) 15:14, 6 November 2012 (UTC)[reply]

http://maf.mozdev.org/ ¦

Reisio (talk) 16:43, 6 November 2012 (UTC)[reply

]

I don't see MAFF or MHT in Google_Drive#Supported_file_formats. Saving the page and putting it in a zip file seems the easiest solution. Or convert to pdf maybe? Maybe someone who uses google drive can help? Ssscienccce (talk) 18:18, 7 November 2012 (UTC)[reply]

Actually for Google Drive viewer (or google docs?), and not Google Drive itself, FWIW. ¦

Reisio (talk) 22:44, 7 November 2012 (UTC)[reply

]

Reisio, I don't really understand what you've pointed to. Ssscience, I'd happily pdf a web page as that would give the single file I'm looking for - how do I pdf a web page? --Dweller (talk) 23:21, 11 November 2012 (UTC)[reply]

Hi! I'm having some difficulty stacking two pictures that I took on top of each other in GIMP 2.4.7 I want to be able to erase some of the top picture so that the next one in the stack shows through. Oh, and I'm running the program on a Linux system. Thanks,

File > Open as Layers. You might also need to try right-clicking on the specific layer/s in the (typically) right-side 'Layers' pane, and selecting 'Add Alpha Channel'. ¦

Reisio (talk) 21:27, 6 November 2012 (UTC)[reply

]

Okay, I'll try that. Thank you, --

talk) 22:29, 6 November 2012 (UTC)[reply

]

The really simple way is to drag an icon of the second picture from your file manager and drop it on top of the first image -- this will automatically add the second image as a new layer. Looie496 (talk) 23:00, 6 November 2012 (UTC)[reply]