Ever coded all weekend?
Miscellaneous Forums/General Discussion/Ever coded all weekend?
Like 99% of us here, I work full-time in a non-programming role. (I do work for a software company though).
I don't get much time in for programming as I'm normally at the gym, seeing the relatives or doing household chores (or playing games) in the evenings.
I'm thinkin og coding all weekend to try to acheive "something" as my projects never seem to go anywhere.
When was the last time you did nothing but code all weekend? Have you ever started coding on Friday night and not slept till early sunday morning?
When was the last time you did nothing but code all weekend? Have you ever started coding on Friday night and not slept till early sunday morning?
I program all day every day!
My day job is as a programmer, although it's very different to the stuff I do as a hobby. For a job I write telemetry analysis stuff and am mostly php based, but my passion is making stuff in 3D. So at some point mid-late afternoon I switch from doing trajectories to vertices :)
The trick to finishing a project is to stop having new ideas, or if you do, to save them for a future version. Stick to the game plan!
My record was 6 days solid, with perhaps 1-2 hours kip a night. Not healthy!
I called it a week when i woke up with my face on the keyboard, a file full of rubbish and the buffer screaming at me. (This only usually happens after many many beers). I also started having mild hallucinations like going to answer the phone that wasnt ringing.
Funny you should say that - I done that last weekend coding in PHP.
I used to have a problem with spending all weekend coding - get home from work (at the time was not doing programming as a job) and start programming from 7pm until ~3am, followed by an early start Saturday programming again from 7-8am until 1am followed by grumpy day Sunday bemoaning lost sleep.
In the end I had to take my dev machine (also my only machine) around to my parents' place so that any time spent on the PC is for a short time on limited occasions. Don't get as much done in the 'virtual world' but get a lot more done in the 'real world'.
My record was 6 days solid, with perhaps 1-2 hours kip a night. Not healthy!
I called it a week when i woke up with my face on the keyboard, a file full of rubbish and the buffer screaming at me. (This only usually happens after many many beers). I also started having mild hallucinations like going to answer the phone that wasnt ringing.
Haha! Classic!
I think the closest I have come is several days and nights..
When I worked in motion capture some years ago, I once worked 27 hours straight - equivalent of three and a half shifts continuously - 9am until 12pm the following day.
I must have driven home because the car was parked outside the house, but the first thing I knew was waking up at 4:30pm at home and not knowing where I was.
It taught me that overdoing it is not a good idea.
I don't recommend not sleeping. It really messes you up for quite a while. So it's not really worth it.
Yes. More hours = more productivity might seem logical, but its actually counter-productive. I needed a day off to recover.
Although overdoing it may be necessary to meet deadlines.
Although overdoing it may be necessary to meet deadlines.
That's the only reason I worked ~28hrs straight once. Would never do it out of choice.
Although overdoing it may be necessary to meet deadlines.
No deadline is worth your health.
And, one thing I noticed while working at Gremlin/Infogrames - the people from 'upstairs' who tell you to work overtime, always disappear at 5:30pm on the dot.
If they aren't prepared to do it, it isn't right that they expect everybody else to. I used to start at 9:30am (sometimes 7am) and I don't recall ever being home before 11:30pm.
I did 65hrs straight once, then fell asleep at my desk. I was young then though, longest recently was about 34hrs during a crunch.
I've another crunch looming too... :/
I once fell asleep on a table, too. After only a few hours of 'work'.
Said 'work' involved drinking pints of beer in a pre-christmas binge-fest. Some bloke started talkin bout one of the girls I was with behind her back, so I gave him a pasting - didn't get in bother cos it was a lock-in ;). The bloke whose face I'd punched turned out to be the brother of one of the girls and I ended up buying him a pint.
I fell asleep on the table and when the landlord woke me up, everybody had gone and I'd missed my taxi by over an hour. Had to walk 8 miles home at 3 in the morning. In proper shoes as well, cos we'd been to a night club that doesn't like trainers. Oh, and the next day my right fore-arm looked like a test card with all the bruise colours that came out.
Bit of a crap night, really.
OK its not really work, but its a suitable anecdote.
Like 99% of us here, I work full-time in a non-programming role.
I don't think that number is even remotely accurate.
Have you ever started coding on Friday night and not slept till early sunday morning?
No. First of all, I'm in the alleged 1% who does do software development for a living. Second of all studies have shown that any programming done above on average 40 hours a week, is effectively wasted, as the result is such poor source code, it will have to be rewritten anyway. With that in mind, and considering that actual programming takes about 20 hours of my 37 hour a week job, that leaves roughly 20 hours left for spare time programming, which is usually done in evenings, as having a weekend with nothing else to do, happens about once in a blue moon.
Although overdoing it may be necessary to meet deadlines.
If that's true, you need better planning. Again anyone who's been through crunch time will attest to this: While you do work for longer periods of time than on a normal schedule, the actual amount of elegant, usable and maintainable code, is dramatically reduced, because as you start getting tired you start making more and more stupid mistakes, and the worst part is they may not get caught in time.
If that's true, you need better planning. Again anyone who's been through crunch time will attest to this: While you do work for longer periods of time than on a normal schedule, the actual amount of elegant, usable and maintainable code, is dramatically reduced, because as you start getting tired you start making more and more stupid mistakes, and the worst part is they may not get caught in time.
Yup. Sadly the programmers don't get to program the all-important development schedule though! In my, admittedly fairly limited commercial experience as a programmer, i've found that crunch time is determined arbitarily by a deadline plucked from the air by commercial needs irrespective of how much work needs to be done by somebody who designs web pages in MS Word.
I don't mind though, I love it :).
Not anymore. I find breaking up focus and doing other things helps me think about what it is exactly i'm going to code (Method-wise).
I also find that I don't lose any progress doing this. Sitting there staring at the background of an IDE like a zombie is less practical then letting your subconcious do the work while you're getting something else done (sleep? beer?).
Like 99% of us here, I work full-time in a non-programming role. (I do work for a software company though).
Count me in on the 1% too. I actually work in a software architectural role, which makes me one who programs AND tells people what they should be working on and how late they should stay to do it. I think I've required my guys to work beyond 40 hours one time, and I was sitting right there with them doing the coding.
Yup. Sadly the programmers don't get to program the all-important development schedule though! In my, admittedly fairly limited commercial experience as a programmer, i've found that crunch time is determined arbitarily by a deadline plucked from the air by commercial needs irrespective of how much work needs to be done by somebody who designs web pages in MS Word.
Again, as Flameduck said, this is a result of poor planning, naive management and/or lack of experience to quantify how much time it's going to take to develop a section of code that solves the problem you need to solve. Commercial needs are generally the need for a product that works. The approach you've described doesn't accomplish that.
However, I understand that you probably don't have a lot of say in it. Just don't get roped into the mindset that "it is what it is" because of the poor decisions your managers make.
The most common flaw in management style is when a manager attempts to negotiate development times with experienced developers. If a competent developer says it's going to be late, it's going to be late. The better course of action is to drop/suspend requirements or consider extending the deadline.
my current working situation is 7 days of 0600-1800 on average, i have orders that have to be on a van at 0600 and would take me aprox 2 hours to pick so i HAVE to get them ready in the evening
now on the 2 days that im off another bloke works for 2 days (or just one if he swaps with another) and can sometimes finish before me (but will a higher error rate by all accounts) my supervisors do meantion this time fact quite often
my only reply
if hes so fast, stick him in there for my days and ill go in on his days off
but they wont
why
cos he doesnt want to do it 7 days in a row
and the main reason he is faster for that one day is that the human body shouldnt be working in 3 degrees C for 12 hours with only 45 mins break a day for 7 days, if he did hed probably be just as fatigued as i am come my night off where i fall asleep with a DVD at 8pm and wake up gone noon on average
ok this has nothing to do with programming but what the mind and body can and cannot do after so many hours and every mistake in code could equate to 4 times the man hours fixing it
Ever coded all weekend?
Right there, right now...
I managed to stay awake for an entire week a few years ago, it makes you rather Ill and I was unable to do simple tasks after for a while.
@Ginger Tea, I empathize with your sentiment, yesterday I was implementing changes in some (baulk) C code yesterday - basically it was coding in C for roughly 10 hours solid - with 15 mins off to eat a kebab with chilli sauce around noon. I did get WELL fatigued as I was implementing more effiecine tlogic in alot of places that was spread over header files and blah.... anyways, after 9+ hours of modifying files on a linux box, I'd to transfer the files to a sun box for the compile - and this amazed me, I only got 2 compile errors, over roughly 20 files in different directories - both easy fixes. I was a little pleased. But yes, productivity is hard to maintain consistently for protracted periods of time.
ive never understood this code for hours mentality that is imposed on people (and elected in some cases)
perhaps thats why most game are patched to hell, if coders had a good work life ballance games might be a bit behind arbitary milestones but the end result might not be as buggy
me i find outside of supermarkets everywhere i want to go to buy stuff (especially regarding my bike) shut by the time i finish so my 'weekends' are filled with all the things i cant do the other 7 days i was at work, like shopping, seeing that movie that isnt on after 7pm, setting aside a whole weekend to just sit and code just doesnt sit with me, not when i have other things to sort out too
In general I'm with you Ginger, but working really hard on a game for some days could be very satisfying too. You should try it some time, I think most people can spare two or three weekends a year.
Some months ago there was a contest in the german blitz community to create a game limited to 10 kB of source code.
There were only 5 days to the deadline as I decided to participate :-)
Using 2 days vacation I got Thursday to Sunday and Wednesday night. So I worked up to 9 hours a day which doesn't sound so much but I need my sleep and some hours of sun and social time.
It wasn't programming only but modelling, texturing and working on sound and tuning AI waypoints.
Half an hour before deadline the AI for my Formula 1 game was finished. I tested it and damn, those cars are fast. So I used the last minutes to implement a difficulty modifier...
And yeah I feeled really good and satisfied after uploading my entry(and the days after). It's great fun to greate a game in a short and clear time-frame and see what you can achieve in this time(and within 10 kB of source code).
The lack in development time is obviously(like that 5 minute title screen), but hey it's a working game. See for yourself:
Download here!(ca. 2 MB)
for a contest i can understand
but for a publisher to say "this game must be out in 3 weeks time" for no other reason than thats the date they slapped on the posters without ever checking the feasability of it being ready
if you had a month in which to do that game you might spend the same ammount of man hours on it or perhaps a few here and there fine tuning it, i doubt you would have committed a whole working month on a contest program if you had work/real life issues to attend to, but knowing you can slap out that kinda work in such a small time frame proves that
A> blitz is great for this kinda thing
B> you CAN do stuff like this if you put your mind to it
Ever coded all weekend?
of course! The real question should be "have you ever coded on Christmas Day and Boxing Day? (more than once)" ;-)
I code every day. Been doing it for about 20 years.
(How come I still stink at it? ;))
I code everyday on visual Studio and BlitzMax, and I also work as a PL/1 freelance, so I can decide the hours I make every day. Usually not more than 7-8 hours a day. Doing more hours would mean lower quality code.
I'm on crunch time. Averaging 18 hours a day for the past 2 months. Will be that same way for the next 2 months.
@BIG BUG: That is by far the most playable game of that genre i've ever played - it's brilliant!
If you dont want to add multi-player code to it, can I? :)