Hacker Newsnew | past | comments | ask | show | jobs | submit | yzhou's commentslogin

minimum 5.5mm^2 is required, so we are talking about 3575 euros here. Besides, it is very rare that you can get a production grade chip on the first try, bug happens all the time, you might need 2 or 3 or even more tries to get a working product.


You might, but there are plenty of examples of first Silicon working, the trick is to spend a lot of time on getting your simulations right.


why do you say Scott Adams is purely evil?


the devil is in the DAC and ADC. You just can't turn 16bit/24bit data directly to/from analog without much loss. The 1/65536 accuracy voltage divider simply don't exist.

So, you have to up-sample the songs to high rates with less bits, like 1bit to 6bits, then do the conversion, and get the best SNR you can.

In this sense, there's simply a lot of advantage of using 24/196 since the above conversion can result in less loss and higher SNR


why was this down-voted too? it's the truth!


There's a big difference in impulse response with different sample rates, any one can see it on a oscilloscope, I bet some one can hear the difference.

Those who don't have a oscilloscope can see the picture here: http://i.imgur.com/wY0wzcW.png


What you are showing is _precisely_ the effect of low-passing, nothing more, nothing less.

See the digital media primer 2 for more information on that: https://wiki.xiph.org/Videos/Digital_Show_and_Tell

If humans were able to hear audio above 22kHz (or what not) in any meaningful way, we'd expect to be be able to demonstrate that effect in carefully controlled studied and then that lack of low-passing may matter; but that isn't what the best evidence so far shows.


The low-passing with a brick wall filter on 44.1KHz audio can be a bad thing sometimes, for example, pre-echo https://en.wikipedia.org/wiki/Pre-echo You won't hear the pre-echo on a 2.8MHz DSD audio.


In the real world, it is almost impossible to make a voltage divider with 24bit resolution. So all the DAC makers have to convert 24bit audio into lower bits(6bits to 1bits), this step requires oversampling the original audio. It is a lot easier to oversample a 192KHz/24bit audio than a 44.1K/24bit audio, and the ringing is much less after oversampling the 192KHz/24bit audio.


The two pictures don't have the same vertical scaling, and it's clear that the probe is ahead of the LPF in the signal chain.


the probe is placed on the headphone jack. The difference in the vertical scaling is that 0dB DSD signal is 6dB below a 0dB PCM.


The brick wall filters used on low sample rate sound cause ringing in the time domain, which can "blur" the neighboring impulse.


the point of using number in the domain is, a lot of Chinese people don't know letters, they don't know English words and they can't get "PinYin"(using letters to represent Chinese words) right. How do you make these people type urls in the browser? well, luckily, they all know numbers!


Are there really computer-literate Mandarin speakers in China who don't know pinyin? How do they input text? The only serious non-pinyin IMEs I'm aware of are the bopomofo ones for traditional characters in Taiwan.


Mandarin speakers in some parts of China speak their Mandarin differently, so can't remember the exact pinyin. E.g. around Wuhan people can't remember whether a word ends with -ang or -an, similarly with -eng and -en. As a foreigner, I can't remember nu and nv, similarly lu and lv. Beijingers have trouble with -r endings.


I've regularly seen people use the handwriting IME in China. They were probably only post-PC computer literate, though :)


many older people like my dad, don't know pinyin well. he doesn't input texts, but he browses web sites. He uses a default home pages full of links to start exploring the web.


Here's what I want: A cheap text ssh terminal with wifi,or cellular, nice keyboard hardware, with extremely long battery life (or solar powered), which i can just throw it in my car and forget it. Whenever I am away of my computer I can always log in to my cloud server and write codes or do some quick fixes.


This should be possible with a Kindle. The basic Kindle costs $50 or so. And it is a fork of Android. And has a month of battery life. There are some Kindle-roots [Search "root a kindle"]. Would something like this work for you? The keyboard will still be the kindle-keyboard.

The other option is a cheap ipad with a bluetooth keyboard. If you turn off all the unnecessary apps and notifications and put it on airplane mode - it should last a couple of weeks. You turn it on only when you need it. You will still need to get an ssh app [many available].


The basic Kindle and Kindle Paperwhite don't use Android. It's only the higher end models like Kindle Fire that use the Android fork.


At least one developer dumped his MacBook and switched to an iPad + Linode server for at least a year:

http://yieldthought.com/post/31857050698/ipad-linode-1-year-...


I have been wondering if Chromebooks could be useful for this purpose...


Could you use a (smart) phone?


Most modern smartphones do not hace physical keyboard, and my Kindle has spoiled me to think that anything less than a month is not “long battery life”.


That's my dream.


SEEKING WORK Remote (SF Bay area) Speciality: C/C++>20years experience , Python, PHP, javascript, digital/analog circuit design, embedded programming, FPGA Contact: jerryzy@gmail.com


I've been using taobao intensively for our business for quite a long time. we are a hardware start-up, we bought components from taobao, make, assemble PCB prototypes from taobao, It's so much better than ebay, for example, they have a IM client for you to chat with the seller, negotiate price and shipping cost in real time, and once the deal is made, they can change the price instantly for you. There also sites that buy taobao things for you and ship them to the US, so all you need is a paypal account or a credit card, and the ability to read Chinese.


taobao also allows you to list contact information like qq, and cell phone numbers, whereas on ebay's totalitarian platform, you cant do this


we have made a A3P250 FPGA board, with build-in USB interface, total cost of building the board is less than $50, can do a ton of things MCU can not do. would anyone want it if we commercialize it? http://cdn.shopify.com/s/files/1/0117/1482/products/IMG_0305... The design was based on http://skywired.net/blog/projects/a3pn250-fpga-breakout-boar..., added regulator chips, got rid of the power-on-reset chips.


If you're not sure, you could start a Kickstarter. That's what it's for - products which are only viable if a certain number of people put down cash.


How does the actel toolkit compare with the xilinx toolkit?


This kind of work is OK for some simple block level circuit designs. In the real world designs, as circuits grow complicated, and parasitics add in, the SPICE convergence issue kicks in and automation becomes difficult.


In a way, do all these different side effects make automation even more important? Modeling them doesn't sound easy, but I would imagine that our mental capacity for evaluating solutions is the real bottleneck here.


I think even if you don't account for all of the side effects, you can still use these tools to develop useful circuits which you can then use to design real-world circuits.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: