christoi

(this post is directed to to the OP) While that isn't too bad of an idea, do keep in mind that the strength of a password strongly relies on the secrecy of the method. To crack any of your passwords, instead of brute-forcing my way through the long hash, I could cycle through works similarly to a dictionary attack, but hash them using the method you used, then rehash them using the site's hashing method. If the method becomes widespread enough, a password cracker could even set up an automated check for that, which would also affect anyone else using the plugin. One possible solution, would be to implement a master password alongside the individual one, a bit like a salted hash (but not as good, since you are reusing it). A sufficiently secure master password made up of random keys would significantly increase the strength of the password without being too hard to implement), plus it reduces rainbow table attacks. I wouldn't say it's as good as a completely random string of characters, but that is a pretty nice compromise (make sure to use a slow hash though, like PBKDF2). Edit: My bad, didn't see that you had already considered that. Since you are using your own secret hashing "algorithm," I suppose that's about as secure as it get's then. I personally don't think I would use it though, since there are times when I need access to accounts on public computers and such (plus for forum accounts I use a password manager, which is also fairly secure).

Both the FX9370 and FX9590 are cherry-picked FX8350's. Overclocked, there's really not much difference between the three chips, unless you are planning on breaking records with LN2. Just stick with the FX8350, spend the extra money on a decent motherboard and cooler, it'll make a larger difference with overclocking. As for benchmarks, see here: http://www.hardwarecanucks.com/forum/hardware-canucks-reviews/62166-amd-fx-9590-review-piledriver-5ghz.html

Couldn't find enough information to say for sure, but I'm rather skeptical of them. Searching 'DK-455' lead me to this page, which shows the box label. Despite the name it's actually a 400w power supply, but the 12v rail can only supply 360w. 3.3v and 5.5v however are very strong with a combined 130w, which suggests it's using an old design. This page appears to state it's actually rated with a 400w peak power and 350w continuous, which is extremely worrying as it appears to be a 450w model rebranded as a 400w (rest of the article seems to say next to nothing, since the pictures are all down). Oddly enough however, Ducky claims it has an 80+ bronze certification, uses Japanese caps only and offers a 5-year warranty. That sounds waaay too good for a unit like that and to be honest, Can't say for sure since there's so little information, but I don't think they are worth bothering with.

To answer the question in the title, more or less. Lower-end phones can easily be found for less than $100, so I suppose there's just not much need for a dedicated player. I've also been looking around myself, but there really isn't many options - especially when you exclude Apple devices. The Sandisk Sansa Clip+ seems to be pretty decent though, from what I've heard. It's very basic in terms of the GUI (even more than the fuze) but sound quality is said to be decent, plus it supports lossless formats like FLAC. Sony also offers a few MP3 players too (NWZE424 and NWZB172), but they lack expandable storage and IMO aren't really worth the price. Though as said, you can sync Apple devices with other programs. You will have to convert all your FLAC files to ALAC, but it's always an option.

For a power supply calculator, I would strongly recommend PSUCalc instead. It's developed by Phaedrus from OCN (along with FiX, who programmed it) and unlike other calculators, it doesn't overestimate wattages. While the outervision calculator isn't too bad, I still find the recommend wattages are higher than it actually needs to be, mainly because the programmed power consumption values are so high (a HDD for example, rarely pulls more than 20w).

There's really no need to upgrade, the 550w unit would be fine. Even with a fair overclock, you are probably looking at an average power consumption of ~400w at most under load.

No, it isn't. I would recommend a minimum of a 650w unit, preferably a 750w. Though why are you making this thread now? Things would be quite a bit different by the time it's next year, with newer cards, different prices, etc.

General rule of thumb is to aim for an average load level of 50%, as detailed here. That usually means for a single-GPU gaming-oriented setup a 450-550w unit is sufficient, 750w for dual cards. Also, brands are almost meaningless with power supplies, since most units are just rebranded. General indicators you can find from reviews are low ripple, high component quality (such as high-quality caps) and good voltage regulation. Anything with ±40mV ripple, ±3% regulation and Japanese caps (teapo is also ok) is what I consider to be very good, but that's only in general terms. A lot of it is really a per-case basis, so you'll have to look into some reviews yourself and compare prices. Also when looking through reviews, stick with sites that actually know how to test power supplies.

With a dual-card setup, each card is essentially rendering the alternate frame. There tends however to be a slight variance in speed between the two cards, so one may be pushing more frames than the second, causing problems with syncing and causing micro-stuttering. At least, at a basic level. There's a couple more complications involved depending on the implementation, but that's essentially the cause.

Correlation doesn't imply causation. There are just certain components you need to improve efficiency (mainly transistors with higher efficiencies) and the use of specific designs. Other things like component quality and ripple suppression are more dependent on other factors, which can be completely unrelated to efficiency improvements. It just happens that manufacturers tend to make their higher-efficiency units higher quality, so they can price the units according to the certifications. A good example is the Seasonic SS-***AT series. It's a bronze-rated unit, but it's actually higher quality compared to certain gold units, such as the FSP Aurum and Silverstone Gold Evolution (at least, in my opinion).

It doesn't, but you can use a molex to PCI-e adapter, which usually is included with the card. Given the rail distribution I don't think it'll cause any issues, capacity-wise it's fine anyways.

Should be ok, the 760's don't seem to draw too much power. According to this average power consumption is around 150w, only peaking to 170w under max load. That means the entire system would average around 320-350w at most, which is fine. Performance of the unit seems to be fairly decent too, so that shouldn't be an issue either. That said, replacing it may not necessarily be a bad idea. The components used with that unit are pretty low-quality overall, so I wouldn't expect it to have that long of a lifespan.

This article explains it very well: http://www.overclock.net/t/973918/a-clarification-of-phases-also-re-exploding-gtx590s/0_40 If you want some more info about motherboard VRM's, you may also want to watch videos. The quality of a VRM is based on a lot more than just the phase count (as the article states), so the video covers other aspects that are important. Not all of it applies for Haswell boards though, since it's quite a bit different.

Not really, though it depends on how you define it. A GPU is very similar to a CPU as it's a chip that can execute set instructions, but it's optimized specifically for specific tasks. It acts essentially as an additional chip to the system, for graphics-related stuff.

Not sure why everyone is recommending 800w units, but a 650w would be plenty for a system like that. The SSR-650RM is a good choice, another being the Antec HGC620M if you want a cheaper alternative.