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*

Friday, February 27, 2009

LOL @ PC donation drive

Minimum Hardware Requirements:

Processors: Original release date should be less than 6 years ago.
Motherboards: Original release date should be less than 4 years ago. ATX form factor or smaller derivatives (e.g. micro-ATX, mini-ATX)
Memory: DDR or newer. Each module should be 256MB or larger. No notebook modules (SO-DIMM) please.
Hard Disk Drive: Minimum size 40GB, IDE or SATA.
Graphics Card: AGP or PCIe bus
Network/Sound Card: PCI or PCIe bus
Optical Drive: At least CD-ROM, read speed 24x and above. CD-RW, DVD-ROM, DVD-RW drives also welcome.
Monitor: At least 15", able to display 24/32-bit (true) color. DVI/VGA interface. Please also provide power and DVI/VGA cables.
Speakers: In pairs please. Nobody wants to receive just a subwoofer or one satellite.
Power Supply: Minimum 350w. Preferably with power cable.
Keyboard/Mouse: PS/2 or USB connector.
Casing: ATX form factor or smaller derivatives (e.g. micro-ATX, mini-ATX)
Software: We would greatly appreciate it if anyone is kind enough to donate Operating Systems, Productivity (office) or any other original software whose license is still valid for installation on another PC. Software should not have been released more than 4 years ago. See additional notes below.

Lets see...

My mom is using a Pentium 4 1.8A for her work - system bought in mid-2002, most likely released earlier - fail
Motherboard - bought with CPU almost 7 years ago - fail
Memory - SDRAM - fail
Power supply - 200+W - fail
Software - Windows 98 + Office 97 - uber fail

She's still freaking using it today. And it still functions as fast.

As for me...

Motherboard - came with Starhub promotion Compaq PC in 2005, mainboard series launched in early 2005 - fail
Graphics card - before I upgraded it to HD3850 2 months ago it was onboard - neither AGP nor PCIe - fail

And the TV rig before it was recently upgraded:

Processor - Athlon 800mhz - released since god knows when, probably 2000 - fail
Motherboard - the famous A7V133 - confirm fail
Memory - either 64 or 128MB, definitely not 256 - fail
Hard Disk Drive - 4GB - fail by a factor of 10
No optical drive, no monitor (uses TV), no keyboard (yes no KB), no casing (yes no casing) - fail
Win98 - you expect me to run slow XP? - fail

Even after the upgrade it still fails - because it uses the exact same motherboard + cpu in my current Starhub Compaq rig. And it uses the onboard graphics. Plus the PSU fails too.

Secondary com - used as dad's com, backup, botting com
Processor - Celeron D 2.4Ghz, may have passed the age requirement, but performance doesn't match that of the 1.8A
Motherboard - some wierd SIS or VIA chipset supporting up to 533MHz FSB - fail
Graphics - f-king slow on-board - fail

Then, at my new job, we're supposed to use laptops in our work. And I suspect the laptop uses Willamette core - fail

And I know of many friends who are still using Pentium 3's or have used it until now. Some of their parents use the same computer for work. And I also see lots of old systems still being used in workplaces.

Those systems with CPUs that fail, they're still being regularly used for real productive work (apart from my decommissioned TV rig (code name Revival) which was used for entertainment). Those systems with CPUs that pass, fail anyway in other requirements, but they are still being useful. My TV rig handles HD 720P H.264 decoding (with slight difficulty sometimes) which is still a CPU-demanding task today, and my main system (which specs are similar to the TV rig less the recent gfx card upgrade) has served me well for years and is going to continue so, and with a graphics card I play UT3 and F.E.A.R. and C&C3 etc with no lag. Even without one it can handle most MMO games with ease, and allow me to frag noobs in UT2K4 under the name of Higu. If you remember this name, yes you're right, you've been fragged by an on-board gfx.

All this fun and usefulness on "failed" systems. Wow cool.

Donating PC to low-income families is good. But the strict requirements makes this hypocritical. What is the use of a high-end computer for children anyway? Even DotA and Maple and CS don't require such specs; my mom's "failed" system, once equipped with a Ti4200, managed UT2k4 and C&C Generals with no lag. And the use of older computers for work illustrates the situation even better.

So lets just say we gave them the good specs to play games, this is the most reasonable explanation since games are the most demanding apps for a PC for a normal consumer.

Now that's just wierd. And you'd think that in low-income families the parents wouldn't want their children to study hard to break out of that poverty cycle.

And you cannot say that I'm too cynical and selfish. Remember, my main system failed the requirements. Even though my systems are all good enough for so many shitz lol.

The reason they want <4-year-old motherboards and <6-year-old processors might be reliability; it'd reflect badly on their image if the donated PC spoils. But hay, reliability and age isn't directly related. My com ran almost 24/7 since it came to me. My mom's rig used to belong to me, and it also ran 24/7 before for a good amount of time. Freak, the fact that all these coms are al still working says that they're reliable enough. Yet if you go on forums, you see lots of people with things dying here and there. I'm guessing if you don't like and respect your gadgets they die on you. My Creative 64MB MP3 player lasted me from 2002 to 2006, and the 2 pairs of NiMH battery I bought from it died first (Sony... good brand).
And if they want reliability, why 350W PSU? 350W suggests a generic PSU, you see PSUs below 500W outside, nobody goes by multiples of 50W anymore. Now imagine the thing has been used for 1 year already. Prepare for some fireworks.
Yay, the irony, the 200+W PSUs that came with the two Starhub systems I have along with the 300W FSP are heavier and better than those generic 350W. Ever tried running a 9800Pro on 300W?

Or, they might just want to give good coms so that they look good. Or because the beneficiary requested for that much. Either way...

I originally wanted to donate the Athlon away (my mom's idea, independant of this donation drive), I'll tell her about this and see what she thinks lol.

I think I'll just donate it to the salvation army. Even the old Pentium II would be faster than the watever OLPC project, because it can play Unreal Tournament. I'm sure it will be useful to them.

The bottom line...

Why bother donating if the people/group you're donating it do doesn't even use/appreciate it more than yourself?

Oh well, I do have a 2.8C and some AXPs lying around. I can donate those.

Tuesday, February 24, 2009

The six stages of progress in audio:

1. First you find EQ sux because it only makes your $10 speaker sound worse

2. With a better but still largely inadequate $30-80 speaker setup, you find that EQ helps to improve sound quality quite a bit

3. With 200-dollar speakers, plus a basic source, you find that EQ makes things too noisy and artificial

4. $500-2000 system, the source is clean enough to handle a bit of alteration, the speakers are beefy enough to dish out that bit more bass, and you also have a much cleaner equalizer (which is probably non-digital in nature). Heck, they have all these capacitors and varistors that EQ the sound inside the gear. Then there's this "tubyness" that is apparently adjustable.

5. $5000 to money-no-object, fk EQ, it's for pussies. You want bass at 8Hz? I GOT bass at 8Hz. My supertweeter hits 100kHz. 5-way systems FTW.
(Note: This is the point where audiophiles with different beliefs split ways. While many will opt for the scientifically perfect system that does not attempt to defy any law of physics and reproduces sound with utmost precision and accuracy, a siginificant number will opt for aesthetics and/or their beliefs in full-rangers and that crossovers are bad. Hence white smooth iPod speakers that cost $6,000SGD and floorstanders with one driver that cost >$10,000)

6. Back to the studio, where it all first started:
"I think the drummer isn't hitting it hard enuff"
"Lets up the bass so that it is"
"The vocalist sucks, she has too much sibilance"
"...so I decrease the high frequencies a bit... done"
"But her voice is too throaty, not forward enough, did I mention that she sucks?"
"Upped the 4kHz, nasal enough for you?"
"Thanks, that gives us the kawaii-ness of the voice we expect to hear from a 30-year-old"
"While we're at it lets clip the electric guitar more, it's not fierce enough"
"And the piano and strings backup don't feel spacious enough, giv'em some crossfeed"

Ahhhhh, EQ + mixer is like the Photoshop of music.

Monday, February 23, 2009

Random thought of the day

Who cares if it's built like a tank.

Do you know that if a tank falls from the 4th storey the people inside will still die?

Newton's laws, heh.

It's not the hardness of the shell that counts, it's the damping, the cushioning effect of the... erm... cushions inside.

Ever wondered why when a car crashes the most expensive part of the car - engine, is always gone? It killed itself to save you.

So even if your product is like this, it may not survive a four-storey drop. Otherwise, it's quite safe from impact and weight, plus it's a nice paperweight on its own.

Sunday, February 22, 2009

Thursday, February 19, 2009

Lame thought of the day:

TripleFi is called "Triple"Fi because it has three drivers.

So why don't UE (or now Logitech) call the SuperFi 5 "Double"Fi? Or the SuperFi 3 "Single"Fi or just "Fi"? Easier to distinguish the two also.

Random thought of the day (continued from yesterday's):

Continuing from the argument that we "hear more details" because they are louder (or the "loud button" effect),

Does adding a supertweeter make guitars sound nicer because of more accurate high-frequencies, or because the highs are now louder?

Lots of implementations of supertweeters do not feature a crossover, making this a perfect target for today.

Wednesday, February 18, 2009

Random thought of the day (And it's actually audio-related):

There's a fine line between detail and harshness. And the terms analytical and clinical are on that thin line.

I mean, harsh sounds sound louder. As do irritating sounds. Like the baby beside you on the bus or the snoring of your bunkmate in army.

And since harshness occurs in the high frequencies where details also lie, more harshness = more details because you hear more things? Or is it because it has too much detail that the recording and playback devices cannot them properly that they turn into harshness? Which, in turn, becomes part of the detail in the sound?

Or, if I use EQ to illustrate to those that play with it, does increasing anything above 4kHz make the sound more analytical? Or more detailed? Ok, probably not detailed, because the system already sucks, and EQ-ing doesn't make it better. But more harsh? Definitely. More analytical? Hmm...

I'm confused. And I believe there are others out there too. Will some master tell me what the heck is analytical sound? Does it means accurate or something else?

Of course, harshness will never be detail, though detail can be harsh. Percussions especially. But if we were to say that an analytical sound means can hear more detail, I dun see how adding an extra violin or two into an orchestra can be bad. Unless it makes the violin section louder for example, in which case it's doing something negative to the sound, which is self-contradictory to the detailed part.

On the same note, stereo widening/soundstage widening does not increase the amount of detail either. You just happen to hear more of the sounds that you like, and trust me, it doesn't sound anymore detailed than the original if you listened to that one carefully or louder. It does bring out the flaws in the sound however, since they also become louder with the sounds that weren't intended to be louder, and I find it much easier to ABX. I understand, for I'm a user of SRS WOW myself, until the day I get enough table space to properly position my speakers. It's natural human instinct to like the feeling of space, hence people are crazy for karaoke and Bose.

And of course, like all enhancements, you lose out something else in exchange. Did someone mention bass? Yea coz bass is non-directional. And the highs, they are highly-directional (hence also suffering from off-axis response), you may get one type of sound louder and another type softer, depending on base or average frequency, how broadband it is, and how it's spread out between the two channels. Did someone mention bass?

And of course, like all enhancements, there's a limit to how much the sound should be altered and can be altered. While I know reverb is nice it's stupid to only hear the reverb of the song. Also, pumping +12dB to 125Hz will not actually give you +12dB to 125Hz, so stop torturing your Travagans and get some real bookshelves of a change. (okay that was random)

Thursday, February 12, 2009

My DAC is so bad, I can hear any defect in the transport!

Read the original story (and title) here:


BTW this guy is good. I like this site.

This brings up an old but interesting and still relevant finding, that a good system does not make your 128kbps MP3 sound any worse than a cheap system, only better.

Before throwing incorrect theories, lets do an ABX comparison. Wait, make that AB. Because the difference would be made so obvious you'd be deaf to miss the X.

Test 1: $10 multimedia pc speaker vs Creative Inspire 2600 vs Creative i-Trigue 2200 vs JBL CS100 vs Paradigm Atom v.1, playback of 128kbps MP3. Wait, that's more like a ABCDE test. Definitions.

This is like shooting fish in a barrel. Yes, you know I'm not making a test here; I'm making a statement.

Before we even care crap about compression artifects, we'd be caring about the frequency response charts of these speakers. Lets recall how a 128kbps MP3 sounds like. Muffled vocals, distorted highs and lows, lack of space.

With the good system, you get muffled vocals, distorted highs and lows, lack of space.
With the lousy system, you get muffled vocals, distorted highs and lows, lack of space, no bass and/or midbass, and shrilling highs for omake.

Test 2: VIA Vinyl Audio (a.k.a. yet another on-board audio solution), Audigy, NeoMini, Zhaolu D2.5A (various incarnations)

It was during this test which was carried out long ago that I realized the topic being discussed today, that a good DAC actually makes the highly-compressed music sound better, gives it new life. I was still using the CS100 + SRS D4 combo that had served me for more than a year at that point of time, and that ensures any difference is due to a change in DAC. (And the SRS D4 sub with two inputs makes it very easy for direct AB and ABX comparisons)

Granted, the flaws are also more obvious, but that's because the details have all come out. The better quality music benefit more (and makes the 128kbps MP3 easier to tell apart), but the 128kbps definitely benefited. There was more fullness to the music, to say the least.

If I'd use numbers as analogy, the 128kbps MP3 scores 60 with the Audigy, and 62 with the Zhaolu. The 320kbps scores 70 with the Audigy, and 75 with the Zhaolu. The 320kbps improves more than the 128kbps, and the difference between them is now greater, but the 128kbps still improved.
This holds true for the op-amp and cable upgrades that I did, from OPA2604 + $7 cable to OPA-Earth + 1694A.

And this is from a person who cannot tell the difference between a bit-perfect and resampled/kmixed source, and can only detect faint differences between 320kbps MP3 and lossless and barely passing the ABX test (but still a pass at least).

(So now we know, if a person claims he can hear the difference between WASAPI and ASIO, yet do not believe in better digital source hardware, he is likely trying to bluff himeself and/or on the forum so he looks pro, and is most likely going to shy away from ABX testing. Though that's not the main point of today's post.)

The main point is, it's not possible for poor-quality music to sound any worse on a good system than on a bad one. Actually there are already good examples out there. Vintage and car radio systems. Radio is crap source. Yet it sounds so good on those systems.

Strange that I didn't realize this when dealing with the more obvious part: the speakers. Though I definitely did hear an improvement.

So what's with this misunderstanding?

Lets check the part that is correct - a good system makes the flaws more obvious and makes the difference between good and bad source bigger.

Flaws more obvious: translates into -> bad sound

Is that correct? Not always.

So don't be too embarrassed to use your UM2 with an iPod nano. Or Aego M with on-board sound. Heck, Aego M isn't the best out there. If it needs lossless to sound good, what does the $100,000 FM Acoustics system need? Alien UFO audio? And you sure didn't buy an expensive system expecting it to sound worse, right?

Which leads us back to the post title. Your system sounds bad because it sucks. Simple. Or your brain is telling your ear it sucks. Same outcome.

Why it's not wise to compare products with performance close to each other

1. I should buy a E5xxx CPU + Biostar mobo becuz it's cheap
2. I should get the Geforce 8 mobo because it's good and just costs slightly more
3. I should just get an Athlon X2 because it costs the same but it's (was) AMD's mainstream part and performs better than E5200
4. I should get a E8xxx CPU because it's just slightly more ex but has much better performance at stock and even more when o/ced to hell
5. I should get a Phenom X3 because it costs around the same, but 3 cores
6. Fk Phenom, it sucks, Q6600/9350 is the way to go

So somehow from the cheapest dual-core I'm now looking at a quad.

Reverse is true also:

- UE-10 Pro
- triple.fi 10 Pro is almost the same, using same drivers
- super.fi 5 Pro isn't too far triple.fi, but 2 drivers instead of 3
- super.fi 3 sounds okay, it's still a super.fi
- Mylarone X3i has the same sound as UE earphones

Conclusion: X3i is close to a triple.fi 10 Pro (coz it's not custom)

Yeah rite

Because, when we say something is almost like another thing, we often fail to measure "by how much". Even a $65 ATP3 can sound almost as good as a $1K system, and I'm not kidding.
Which is how there are reviews sites like 6sun that can claim a $39 T-amp has similar performance as a $3200 integrated, and online forums like neck-fi where everything seemingly is a poor-man's-version of something else more expensive.

And strangely, my non-audiophile family members can tell the difference between "two products with similar performance" easily.

Have we all become audiophools who listen with our brains instead of our ears?

Sunday, February 8, 2009

You may think I'm mean, proud, egoistic, cynical...

"hmm.. as a under grad engineer =p
i think asus xonar is value for money!
cause it's a DAC
digital to analog converter!
hmm.. cause MP3 is a highly compressed audio..
DAC map the wave form of the audio to the closest possible analog signal..
in a way.. the audio is more real.. not so stiff.. got bends on the signal.. more vibrant

this is the theory part though.. not sure about x-fi dunno what chip lai de
but.. i tested xonar and x-fi
gaming wise..
i see no diff..
maybe x-fi SLIGHTLY better"

But if you read the above statement and truly understand (and I mean TRULY) what's wrong, then you're on the same side as me.

Anyway I think he fail his uni liao. Or he not doing computer/electronic/electrical engineering but mechanical instead.

ADD: Oh cool he even has a graph! Sh!t I really can't stop laughing. The graph is here just go see it yourself. The reason is clear... have you ever seen before a digital waveform?

Dunno if I've said it before, but the reason I blog behind people's back instead of correcting them in their face is because 1) I may make mistakes in cynical statements of mine (of course, that's when my suaning mode is on, against certain products especially), and more relatedly 2) Trying to reason something on forums usually ends up with flame wars, with people who don't believe that they're mistaken and usually with more people joining their side. And especially in audio, where people have different religious beliefs. And pressing one's belief onto others is bad, not to mention it's a kind of trolling. That's why I avoid that. I post my opinion here instead, where less people can see, those seeing won't intervene as though the person being talked about is his dad.