Is this a GPU problem?

Seriously, it really isn't that interesting. Honestly. You would be very disappointed (and bored to tears) if you watched them.

Is it? I honestly have no idea. I know very little about hardware. I'm a software guy.

Gotcha. If I were you, I would just look into what resources are being used when video is being converted. Is CPU maxed out? Are all your cores being used? Is memory maxed out?

I would guess that video conversion is probably way more CPU-intensive than anything else. And, if that's the case, you can see that immediately just by checking your resource usage.

If it's not that, it might be HD speed. I dunno if Amazon offers SSD data storage or what your budget is, but that or holding everything in RAM are the only real solution for HD-dependent run-times.

Thanks man.

It seems to be the consensus answer that its a more CPU intensive process than memory intensive (i guess thats the term? whatever).

Unfortunately Amazon doesn't use SSD storage. That's actually one of the most common gripes about them.

They do, however, offer a a high CPU instance with proportionally more CPU resources than memory - it's got 7GB of memory and 20 EC2 units (8 virtual cores with 2.5 EC2 compute units each).

It isn't terribly expensive - a year long commitment is about $3400. I'm thinking about starting an instance just to do some benchmarking to see exactly how much quicker it works on that beast.

If anyone else has any advice I'd love to hear it.

You ignored my primary piece of advice. Just check the resource usage yourself while it's converting. On linux, this means running the "top" command. On Windows, it means opening task manager.

Sorry, wasn't being clear. I'm definitely doing that first. I was just saying I'm also gonna play around with it a little bit just to see how well it compares.

I'm definitely all about going for the free (and simple) solution, particularly when it's coming out of my own business.

I've never used Amazon but you should be able to just ramp up the RAM quite easily right? That would probably be the easiest way for better performance in converting if you can fit the whole thing into memory. You say it's 7.5gb, try going to 16gb and see if it helps much. Also I think you want as many CPU cores as possible.

Figured it out.

By default, multi threading isn't enabled in ffmpeg. It just has to be enabled on the command line to take advantage of more than one CPU core.

Easy fix.

I have a database table of sports teams and their players. I want to create a web form that will dynamically generate based on user input. For example, the user will select a team, then the second part of the form will generate without refreshing, giving the user choices for the players of that team.

I assume I need to learn AJAX to create this. Correct? Since I need to interact with my database.

Yep, but ajax isn't actually a programming language, though. It's more of a technique where you use javascript on the web page at the user's end to interact with the server without reloading the page.

There are basically going to be three parts to make it work: the web page/javascript on the client side, some sort of script on the server to handle the data coming from the user (commonly PHP or Python or Ruby), and of course the database where the data is stored.

If you're interested in learning about javascript/ajax, I'd recommend going through some javascript tutorials to get the basics down and then checking out jQuery. It's a very well developed, widely used javascript library that makes common tasks like ajax very easy.

Learn about graceful degradation. If you write the page first that works when AJAX is not available it will work for everyone (win).

Then when that's working, add another layer on top which will do fancy AJAX stuff.

This design pattern is really really good for lots of reasons.

Are you using the -thread n switch?

Edit: nevermind

just wanted to pop-in and say: Goddammit, I love Pig/Hadoop! what a joy to work with.

I'm sorry, Hadoop just sounds like hippie technology used by people with dirty feet.

i'd rather eat granola and work with Hadoop, than foie gras and Cosmos. what a cluster****.

I've had a hard time explaining some resume gaps recently. I'm currently unemployed, my last job was as a web communications analyst for nearly four years. Before that however...I was a pro poker player (gasp!) and a college dropout.

I've been skirting the issue, trying my best to close the gaps on the resume with projects I've worked on (even though it wasn't how I made my living).

Have any of you been truthful about your poker past in the world of tech job searches? Should I just come out and say it?

You did no consulting at all during that span?

I have poker pro in my resume for the time where that way my main source of income. My assumption is a simple one, if people don't see how that could actually be an interesting skillset or at least try to understand it it's a workplace I don't want to be stuck at anyways. So it's actually a pretty good filter (if you feel you can pick and choose your employer and are not in "need work" mode).

Despite what the poker coaching industry and so forth want to tell you making a living from poker is pretty hard. You need a very solid mindset and understanding of variance vs your skill degrading. You need to be self motivated and constantly learn and improve. You need to understand how the long run works. You need an excellent analytical mindset.
All that is valuable.

man i just spend TWO DAYS trying to get a demo php app for a web api working. fv<kin thing was full of bugs, php syntax errors, etc. jesus christ.

• inconsistent variable names $productID and $product_id
• incorrectly using variable syntax when hes defined var as a constant instead
• bad links across the project <a href="/billing.php"> seriously "/"
• 'undefined constant' errors cause the guy didn't quote his strings

its like the guy just posted the project on github without ever testing whether it worked or even paying any attention to php error messages

at least now i'll learn how to use github and fork a project. these bastards should be paying me

I'm looking for some ideas on the best way(s) to proceed with a website. I have a lot of data organized in excel files that I want to extract and display. What I'm looking for, I assume, is getting into database management.

Primarily I am looking to recreate the tables from excel to be displayed in a similar fashion. How do I take the cell information and transpose it to the site? This involves (again assuming here) converting the excel to some format (would love to avoid manual data entry) that some language/script will then turn into tables. Is there a way I can just convert to csv, upload, then get output? I can do the css/html for the output, my issue is I don't know the best way to get that output. In terms of size it is ~10 sheets, none being uniform, with anywhere from 20-500 cells of data per sheet.

I realize what I'm asking is very general but I have no experience in this area so I don't know what I should be learning. Any help is appreciated.

Do you need your website to stay in sync with any changes that happen in the spreadsheet? Is the data going to change at all?

Anyone of you hired programmers have any thoughts on QA's? I might be getting a job as a QA for a software company which makes roster programmes. The office employs about 5 QA's and around 10 programmers. I am not very technical at all and know nothing about programming. Is this normally the case with QA's who you have dealt with or will I struggle to interact problems effectively with the programming team with no IT background?

(I think I worded this post awfully, sorry I'm very tired.)

QA is more about being methodical and thorough than necessarily being able to read the code. It'd be good to pick up the architecture so you know where what you are seeing on the application is coming from, but I'd rather have someone who can narrow down the cause of a bug and document it than a wannabe dev.

Yes, but that isn't a major concern at the moment. Eventually I would like it to be so that all I have to do is update the ss, upload/convert it on backend, it gets updated front end.