W A R N I N G !

W A R N I N G !

This page is full of non-facts and bullsh!t, (just like the internet and especially forums and other blogs), please do not believe entirely without exercising your intellect. Any resemblance to real things in reality is purely coincidental. You are free to interpret/misinterpret the content however you like, most likely for entertainment, but in no case is the text written on this blog the absolute truth. The blog owner and Blogger are not responsible for any misunderstanding of ASCII characters as facts. *cough* As I was saying, you are free to interpret however you like. *cough*

Sunday, February 28, 2010

Save on those input capacitors!

What started out as a curiosity turned into a revelation -

What happens when the high-pass-filter of a source is connected to the input capacitor of the receiver? Essentially you get a circuit like this:

So I was wondering about the cut-off frequency. Ran a spice above and guess what, capacitor C2 doesn't have any effect on the frequency response.

Well it makes complete sense, as a capacitor in series will only block current if there is any flowing through it. In this case, there isn't.

Even if we model the input of the amp chip as having an impedance of 1M ohms, even with a 0.1uF capacitor (1/10 value of that above) changes to the frequency curve is minimal.

So, that 1uF capacitor was enough.

If a pair of lovers become one in mind and soul...

...And they love each other...

Doesn't that make them narcissists?

Thursday, February 25, 2010

Geek =/= Good in maths

This post is written in response to Jason Fox of Foxtrot comic strip - a stereotype of a geek that is also good in maths yet somehow do not behave logically.

Notice my use of "also" in the above statement.

Yes, what I want to say is that being a geek does not require you to be good in maths.

What makes a geek one in the first place? The definition of a geek will greatly decide if he is good in maths or not. If by geek you mean a person with a degree in computer science, then hell yea he is good in maths. But if geek you mean one who watches Star Trek and knows how to build a computer, then no.

A geek does not need to be good in maths because his area of interest may not require it. Watching movies or playing video games (although this has a higher tendency to be called "nerd") does not require maths, same goes for assembling computers or rewiring your house's networking cables. This is where technicians come in, and technicians do not need to be good in maths.

Let's see the definition of a geek:

Wikipedia -
a peculiar or otherwise odd person, especially one who is perceived to be overly obsessed with one or more things including those of intellectuality, electronics, etc

Dictionary.com -
1. a computer expert or enthusiast (a term of pride as self-reference, but often considered offensive when used by outsiders.)
2. a peculiar or otherwise dislikable person, esp. one who is perceived to be overly intellectual.
3. a carnival performer who performs sensationally morbid or disgusting acts, as biting off the head of a live chicken.

In neither definitions was being good in maths explicitly mentioned. Computer and electronics are explicitly mentioned though, and the term geek is commonly associated with these.

So being a geek does not mean you're good in maths. In fact, many a times the contrary happens - because a geek is too engrossed with his enthusiasm he neglects studies and scores badly. I do know of many like that. And if one's not good in maths he's not good in logic either.

The same logic can be applied conversely - being good in maths does not make you geek either - see the student council. Ok, maybe they are geeky, but in a different way. And I do know of people good in both academics and sports.

I'm typing this, because I know that I'm a level 0.7 geek, yet I'm taking heavy damages from maths quizzes.

So yes, geek =/= good in maths

I also hate technicians that call themselves engineers - copying product designs from datasheets does not make one an engineer - I can get anyone from ADM to do that, and they bundle it with good aesthetics and photography. (Coincidentally, it just happens that a friend I know studying design is more experienced in soldering and machining than me)

Being an engineering student does mean being good in maths. But you'll still get a 'D' if you're "good". In this course, "excellent" is the average.

It's no wonder that after they come out they become another sub-species of humans. One that as a result of natural selection has great difficulties in procreation, because they are very incompetent in the process of courtship, also known as dating. It is rumored that a huge percentage of them die as virgins.

Wednesday, February 24, 2010

Alternative interpretation of 2012:

The Mayans might have predicted year 2012 to be the end of the world.

My brother commented that they might be just too lazy to continue.

But my interpretation is that they did predict year 2012 to be the end of the world - for them.

In that case then they had grossly overestimated themselves.

Monday, February 22, 2010

G.A. Geijutsuka Art Design Class

I forgot to blog about this underdog of 2 seasons ago.

It's both some sort of a coincidence and not that three painting-related Animes I saw I like all three. (And Canvas doesn't belong in this list) But unlike Sketchbook and Hidamari which are slice-of-life's, G.A. is more lively and has more actual things to do with art - think of daily nonsense that revolve around art/school life. All three share the same commonality of being originated from 4-coma though (that's why they are interesting).

SD characters with well-developed personalities, unsharp but artsy-looking background, no story but lots of plot, no fanservice but who cares.

We have been so focused on the evolved aspects of Anime, this Anime reminded me how animation should be like. Just fun and easy. It also reminded me of the times when I was studying painting/drawing, either parentally-forced or out of self-interest. It's not all fun and easy though. If there's one thing that I hate most about creation (of any form - picture, music, electronics), it's the long waiting period between idea creation and product completion.

This show makes it onto my rewatchable list, hence I'm recommending it. On a related note, Seitokai no Ichizon is also on my rewatchable list, but I guess this show is already well-known. It's great, for the wrong reasons.

Mmm... Echo-Of-Death-niisan...

Sunday, February 21, 2010

$50 per driver repair

...that's what I was quoted by a shop. The other shop I wanted to ask for quote is still on holiday perhaps.

This $50 better includes reconing, though I wouldn't bet on that. $50 just for refoaming...

That's $100 for a pair. Lets see, my Atoms cost me $30 with nonexistent foam surrounds, add $100 for repairs,

I can get okay-condition second-hand better speakers for that money, things like Tannoy, Mission. And only kok-quality or very-old speakers would have deteriorated surrounds. For cheap speakers, $100 is too much. For expensive speakers, would you risk the sound-quality drop with new surrounds that change the speaker's parameters?

Heck, a pair of new good replacement drivers costs only $200-odd. But the only reason I'm not getting them is because this particular Rogers speaker have a f-king squarish mounting that does not fit almost all other drivers. Mounting it from inside is not possible either.

So if I can spend $100 to get these legendary speakers working again it'd be great. But that would require a reconing, at least.

Saturday, February 20, 2010

Asus HD4770 - EAH4770 FORMULA cooling fix

Been a while since I last posted something not related to audio.

This time, I'm ranting on my graphics card - Asus EAH4770 FORMULA.

This crap has been creating a mini tornado inside my computer since the first day I got it. Which is strange because Asus touts better cooling solution with their FORMULA (yes spelt in all caps) series, and the previous card from them I had - a HD3850 - was quite quiet for a high-end card of its time while using the generic Glaciator cooler of Asus.

And to make things worse, Asus confirmed that fan speed control doesn't work on this card - a suspicion I had since the fan is spinning as fast idle and load.

Finally pissed with putting up with the noise, I ripped the card from my system to give it a fan-job.

And this is what I saw -

Trying to cool more than 50W with this heatsink, no wonder the fan is spinning like mad. I've seen cheap CoolerMaster CPU HSFs that are more powerful.

It gives me goosebumps to know that higher-end cards are also using this heatsink.

The solution - a 25mm-thick 92mm fan removed from the chassis during my SPCR days (now made useless by my tinnitus) because it was too loud. It's capable of 2000rpm. Plugged to the motherboard for fan speed control.

However, doing so only proved the need for a powerful (and loud) fan - temperatures reached dangerously close to a hundred in FurMark testing, and that's with the fan at full speed. While it should not go over 80-ish in normal gaming, it's probably not good for the card in the long term.


Apart from overclocking to gain performance, one can do undervolting to reduce power. Power is proportional to voltage squared, so any decrease in voltage is great.

There are two ways to do this - the first involves flashing the card's BIOS to Asus EAH4770 TOP's, which allows voltage adjustments in software. The second is modifying the BIOS itself to run the card at lower voltages by default. Both have their risks - cross-flashing another card's BIOS is never a good idea - a change in just the memory timings information can stop the card from booting, while a voltage set too low can prevent the card from booting too.

Unfortunately/fortunately, the lowest this card can go software or BIOS is 0.9V, 0.1V down from original 1.0V, so there isn't much point for the software voltage control, and 0.9V should be still rather safe for the card to boot, so I went for the safer route.

Even though this is only a 10% reduction in voltage and 20% reduction in power, when things are overheating you need all the help you can get your hands on.

But with the GPU at 0.9V in 3D mode, it crashes frequently (and restored successfully each time by ATi's VPU recovery) at the default clocks of 750MHz, so I cut it back by 10% to 675MHz, all smooth onwards. This shows that graphics cards do not have much headroom compared to CPUs.

With these new settings it is safe to say that the temperatures don't go over 80 in FurMark with the fans at full cooling setting -

Full cooling mode

Fan1 is the 92mm fan cooling the gfx card, Fan2 is the 120mm CPU-cum-system fan - the CPU has no fan of its own, and with a Ninja cooling a heavily-undervolted A64 there is no need for one.

FurMark temperatures

Quiet mode

With normal games and the fans at quiet mode temperatures don't reach 80 either. So mission accomplished.

It's a pity I lose auto-fan-speed-control with this method however.

Friday, February 19, 2010

Audiotrak Prodigy 7.1 HiFi vs Creative Sound Blaster Audigy ES

Another "view" of mine. This time it's Audiotrak Prodigy 7.1 Hifi.

I mentioned before that a review without any basis for comparison is utterly useless. It's akin to masturbation. So I have two opponents against this guy - Zhaolu D3 DAC for analogue output, and Sound Blaster Audigy for digital output.

Prodigy vs Zhaolu D3

Not much of a fight here - the sound from the Prodigy is overly bright and harsh yet still lacking high-frequency detail, as though it was purposely tuned bright to make up for lack of detail. Bass is rather okay in quantity but does not hit low enough, and is boomy and anemic, quite characteristic of the JRC4580 op-amps it's using. Overall it's good for a cheap studio rig, but one deserves better. Replacing the op-amps with LM4562 might make the sound more likable at the expense of tonal accuracy, as I found to be always the case when replacing a 4580/4556/5532 with LM4562. Remember, tonal accuracy is very important in monitoring, much more so than detail, because our ears care more of tone than detail, so do most speakers.

In comparison, the combination of Realtek digital-out + Zhaolu D3 is much more pleasing, showing that the analogue stage is much more important, and it makes more sense to invest in a better analogue stage than split the money evenly between analogue and digital. Of course, the final speakers stage is still the most important.

So Prodigy 7.1 HiFi vs DAC, guess which one I would pick. Of course, my D3 isn't a bad DAC, though I can't say the same for USB DACs in the $150 region.

Prodigy vs Audigy

Prodigy vs Audigy, heh. And they're not by the same company, not even companies on the same continent.

As you can see in the picture above, the Prodigy is one heck of a bling card. With a WM8776, the Nichicon MUSE and OS-CON capacitors and some big ones for power supply, if this doesn't sound good just because it draws its power from the computer, you can be sure that all USB-powered DACs don't either. Still it doesn't beat the Zhaolu, not by a fair margin.

If you think that my judgment would favour the Audigy due to familiarity, well you're wrong, but congrats for being attentive. All this while I've been listening to the Audigy with EQ on to tone down the bass of room acoustics and the brightness of these speakers. The comparison will be made with all settings at default, so both will be newcomers here.

Vs the Prodigy, the Audigy's sound seems to be missing some things at the top end. And the Prodigy has this extra sparkle there.

A sure win? No, time for some AB test.

Out of curiosity I tried ABX 320kbps vs uncompressed again. And failed again. Seems that only with some material you can tell the subtle difference, and this isn't one of them.

Back to Prodigy, AB testing yields no conclusive difference. So only time will tell.

Meantime I say just get a decent soundcard and a DAC and be happy with it.

Playing with scheduled posting:

I got a few more posts coming up, and I'd like to spread them out.

What does this better than scheduled posting.

Thursday, February 18, 2010

Bad progress on Sure-Q.E.D.

When bad things happen, they don't happen alone, they do a 7-hit combo.

Finally managed to solder the soft-start and remove the resistors to ground. First thing that happened when I hooked it up to my main system was a huge-ass-loud high-pitched buzzing. Apparently some bad connection or something, or maybe a short to ground. Got it fixed temporarily by connecting the banana connectors straight to the amp board.

Next, there was some hissing from the amp, but only with the signal wire connected to the amp PCB. EMI from the PSU.

So I used aluminum foil to wrap around the wires and capacitors as shield - why bother spending much money on shielded stuff when something cheap works better. It seems that class-D amps require triple-shielded input wires, I have aluminum foil wrapped around the wires a few times, this should count as 1.5 layer. And I have two of them - one for each individual channel (or each channel + ground where they split), and one for the whole bundle.

There is no potato underneath.

This sure helped a lot, and I haven't tied the drain wire yet, but there is still a tiny bit of hissing left, perhaps the drain wire plus some additional shielding will help. There is also some high-pitched noise when the socket is not connected to a source that is switched on, perhaps some grounding issues. I'll try grounding the signal ground to see what happens.

However, while moving the length-limited wires, or more correctly called bulkhead now, one of the disturbed solder joints on the volume pot came off. So the amp is not working again.

And to make things worse, my iron was spoilt for some reason, it just does not heat up.

It was meant for me to change it since long ago anyway. I've had many bad solders because of it. I think it has a short-circuit too somewhere, as I can feel current flowing through my hand if I hold the solder.

What was supposed to be a soldering iron

A new one, preferably with ceramic tip for long-life, and a trip to DX for the accessories. SLT still sells the good irons and third hands cheaper.

And to top it off, I couldn't check the connectivity of the connectors, or where the iron was broken, because my multimeter was out of battery because I accidentally forgot to switch it off after the previous measurement, and the neighborhood hardware store that has its battery was closed for CNY.

And I couldn't go home to complete this this weekend because I have Maths, Physics and Material Science CA next Monday to Tuesday.

Wednesday, February 17, 2010

Ever heard sound leaking out from others' IEMs?

Well I have.

So much for isolation.

So much for safe hearing too.

Never take words out of context...

Yesterday I was doing some checks on my speakers (after they suffered from a 1.5-foot drop from the top of my computer), when I decided to shift them to the living room to see how they would sound there.

First thing I noticed - there was no bass.

Second thing I noticed - there was no bass coming out from one of them.

Third thing I noticed - I forgot to plug back the internal wires to the woofer for that one when assembling.

After the fix, first thing I noticed - there was more bass. Actually the thing still sounds rather okay, but the onboard Realtek of the TV computer was obviously struggling, crackle and boom everywhere.

Second thing I noticed - the computer restarts itself after a few ten seconds of music playback.

Third thing I noticed - my wireless PCI adapter was undetected again.

Forth thing I noticed - the bass from the speakers rattled the card loose from its slot again.

Fifth thing I noticed - there is bass at 50cm from the TV, but not much left after that.

A living room is I'd say a mid-sized to large room, lots of bass is needed to fill'er up.

These speakers sound full of bass in the corner of my room, yet devoid of it outside.

Here in the living room, bookshelves are bookshelves. And if you need bass, 12-inch bookshelves.

So when some people tell you that bookshelves have as much bass as subwoofers, you have to consider:
1) Is that guy trustworthy or is he just some schizo maniac trying to act useful and function the opposite
2) Does he have working eardrums and use them to listen, or takes placebo effect and ego to the extreme that everything he owns is perfect
3) Is his room a large room and what distance is he from the speakers

Anyway the first one you have to consider every time you read something on internet forums, the second one you consider on audio forums.

In short, bookshelves are best for near-range and bad for far-range.

Tuesday, February 16, 2010

System matching/synergy

I need to suan something now and then. And the subject of today's suaning is system matching in audio setup.

System matching, or synergy in managerese, is a practice/belief that a certain combination of components, even if they are each the best on their own, will create a sound that is not pleasing to hear, and vice-versa.

This begs the question: How would it be possible for a combination of accurate components to create bad sound, since bad sound means inaccuracy?

Well, nobody said expensive stuff must be accurate. In fact, there are many gullible people who put in lots of money and effort to make their sound shift from originally intended, like putting extra capacitors in series where not required.

When people do things like that, the system shifts from being neutral to biased. Which may sound good initially.

Then, another part is introduced into the system. If it counters the previous part and makes the system neutral and sound good again, then system synergy is achieved.

If it does the same as the previous part, then the effect is x2 and the sound is crap - system synergy has not been achieved.

So actually, why bother with system matching - just get whatever is supposed to be accurate and skim on unnecessary tweaks.

Monday, February 8, 2010

Sure-Q.E.B. v1.2b

Sneak peek. The product isn't functional yet due to technical issues.

I skipped v1.0, v1.1, and the version a upgrade and went straight for v1.2b instead. Well, it was more convenient to drill more holes and solder more things in one shot, and some parts arrived earlier than expected to make this possible.

Yes, originally there was supposed to be these versions, and I even planned what upgrades these versions were supposed to have, though they never saw the daylight:

Version number -

v1.0 - everything stock with no upgrades

v1.1 - 1 to 2x mid-size power supply reservoir/tank caps - Panasonic FM 2200uF chosen, 2.2uF film caps for input capacitors - WIMA MKS4 chosen (though on hindsight should've gotten EPCOS MKP, but I was overrun by the WIMA brand, but it don't matter now) - I call this the "critical while making financial sense" upgrade. Yea, many people's upgrades don't make sense, considering how much they spend vs what they get out of it.

v1.2 - 10000-20000uF reservoir cap (15000uF ELNA For Audio chosen) + 2x boutique input caps (3.3uF "Solen OEM" chosen) - this is the "upgrade to match if not beat boutique amps" upgrade - yea, Virtue Audio has a 22000uF Panasonic in their Virtue ONE, I'm going to counter with 15000uF ELNA + 6x330uF already on the board itself.
This is also called the "upgrade even if you can't hear improvement" upgrade, because with the ELNA I need longer leads and higher impedance compared to the 2xFMs, and with an SMPS in the same enclosure I would need less capacitance than a linear supply or an external PSU, so the 2xFMs might be more than enough and in that case give better performance. But this upgrade is all about bragging rights since (certain) numbers are all that (some) people care.

v1.3 - the "all out" mod - what's done in v1.2, + the two 2200uF FMs soldered onto the board closest to the amp chips, maybe bypassed with the now useless WIMA MKS4s. But the highlight of this version would be a complete replacement of the output filter, this would cost me $30-50 and more for the better coils. This will be the finale to make this amp better in all aspects than all commercial amps in the same price range. And I've read that the output filter is the most chui part of this amp (I could've guessed by looking), and upgrading it yields the greatest improvement in sound quality.

And the alphabetical suffixes -

a - a for attenuator, a.k.a. volume control. It cost $20 + many wires + messy soldering + a hole and a reduction in sound quality, but it is required for safety if I want to use this amp. I prefer to call it "volume limiter" instead, and it's stupid to control your volume just through the computer - you'll know why if you have a power amp and your Windows settings are reset or something.

b - b for bling - see the white light inside the amp? A 24V LED strip with 12 white LEDs in series, controlled by a switch. I expected it to be brighter though.

c - c for source changer, probably will not see the light of day - the big aluminum knob might have taken up too much space for an additional switch + input jack on the front. An additional input for 3.5mm is proven to be very useful when your computer is unable to produce sound (like down, reinstallation, or doing something else) and/or connecting to another source e.g. PSP for entertainment or AB test purposes. Yes I could have used the switch of the LED for the source switcher instead. But I decided to go for bling. Because bling is good.

Saturday, February 6, 2010

STADS - change your audio output device in a jiffy

Well, those of you geeky enough to use two soundcards should know how to Google, but I'm going to share this anyway.

When you have two audio devices, and you want to switch the one being used, it's a trip to Control Panel -> Sounds and Audio Devices -> Audio tab -> Sound playback -> Default device

It gets rather cumbersome after the first time.

Fortunately there's always help. With System Tray Audio Device Switcher - or STADS, switching the default audio playback device is as convenient as two clicks at the system tray, just like changing AA/AF settings of your graphics card, and over/underclocking your CPU.

Ok, so I made those easy too through some settings. Nvm if it is still long-winded at your side.

But with the STADS -



And it's done.

And hell yea I got a Prodigy 7.1 HiFi. Why do I have a Prodigy 7.1 HiFi? That'd be covered in another post.

There's no link to the software featured. Cos as I said earlier, those of you geeky enough to use two soundcards should know how to Google.

And damn while typing I thought of something but forgot what it was when typing.

Wednesday, February 3, 2010

The success of Twitter...

...is owe to the human nature.

Surfing forums, blogs, facebook, and chatrooms, it seems to me that many people want to let their opinions be heard yet do not have anything to substantiate their views.

Like, for example, a simple statement of "you suck".

An argument would require elaboration and support. But not a statement of "you suck".

Granted, this means such a two-word sentence is utterly useless and meaningless, but that is what many people like to write.

And they want themselves heard.

And what better place would that be other than Twitter?

And diaries, yes diaries, for the exhibitionists that's what blogging is all about. But some blogs I've seen have their majority of posts a few sentences long. Either lack of expression skills or a boring life. Twitter is a good replacement for blogs.

Screens with narrow viewing angles

Screens with wide viewing angles

Privacy filters

Why do I bother buying a screen with wide viewing angle then paste a privacy filter over it? Shouldn't I just get a screen with narrow viewing angle?

Because there are people with different needs. People who need wide viewing angles, people who need privacy.

And the market isn't big enough for both, so the smaller one becomes an add-on.

Following this argument, "warm-sound" add-ons are there because people want them to be.

And capacitors that redefine physics.

Tuesday, February 2, 2010

Anti bullsh!t radar

I'm okay with China products.

I'm okay with angmoh products too.
(same goes for Japanese)

China products sold by angmohs

Now this I'm so not okay.

China design at angmoh prices.

It's only possible with China:


The world's factory

The source of cheap products

The source of many cheap DIY-ish audio products.

A lot of China-originated audio equipment look as though they can be made from ground up by an individual. Kit enclosures, through-hole parts, wires connecting sockets to PCB, I won't be surprised if even the holes are manually drilled.

Apart from the obvious lack of R&D (how often do you see SMPS, discrete output stage, programmable microcontroller in these devices?), these often small-scale makers don't have the economics of scale to have an automated production line. So no SMD parts also - many companies claim through-hole components and hand-soldering are better, but these are just excuses for having to use humans for production, I've seen soldering jobs worse than mine more than once, and I'm a poor solderer with my POS iron.

Or, is it because labour is so cheap in China, that everyone choose to optimize their production for humans? Hence lots of those parts for DIY are being made. No chassis made exclusively for any product; everybody use similar boxes here and there. Try to recall a Chinese product that isn't industrial like.

On the good side, lots of makers can build products from scratch. On the bad side, lots of makers are building their products from scratch, and there's a limit to how much individuals can do.

So we'll never see a Chinese DAC that costs over SGD$500 (China price) without some sort of cheat factor involved (like tube, for example). Actually, not much choice exists at over 2500RMB anyway, and only a few have a relatively long history of making DACs. Even so, without at least 10 years of experience in design, it's hard to demand too much money.