Topic: source data for mtgox depth (Read 1551 times)
I have a similar problem. I want to create a very simple app that displays the 20 most recent bids and asks im trying to fiddle about with the websocket API mtgox has but its a hassle as i know very little websocket programming. I was also wondering whether theres some neat way of only requesting the x most recent bids and asks or something of the sort. I am using python for this project. Thanks in advance for any help!
You're welcome. What programming language do you want to use? Maybe I have some code fragments for you to start. 
Thank you dexX7 for pointing me streaming socket.
Didn't knew about that order book can be grabbed from there.
I haven't used streaming socket at all, so far only url api calls.
It is definitely the way to go in such case.
As of now I will not try it, have some other things to do now, and handling streaming sockets in R environment might take some time.
Thanks for the answer!
Didn't knew about that order book can be grabbed from there.
I haven't used streaming socket at all, so far only url api calls.
It is definitely the way to go in such case.
As of now I will not try it, have some other things to do now, and handling streaming sockets in R environment might take some time.
Thanks for the answer!
Sorry, I'm not sure, if I got your question right, but you want to maintain a live orderbook without rate limitations? You don't need to download the whole book every few seconds.
I think this is the way to go:
- download full order book
- subscribe to order book updates via their streaming socket
Because you only receive updates, traffic is quite low, but as rate limit allows, you can and should do periodical synchronization by downloading the full book.
I'm not sure about the reasons for not downloading the full book, so I can only guess that it's about rate limit and data transfer.
Does that satisfy your needs?
I think this is the way to go:
- download full order book
- subscribe to order book updates via their streaming socket
Because you only receive updates, traffic is quite low, but as rate limit allows, you can and should do periodical synchronization by downloading the full book.
I'm not sure about the reasons for not downloading the full book, so I can only guess that it's about rate limit and data transfer.
Does that satisfy your needs?
anybody?
Hello,
mtgox already provide api for truncated market depth: https://en.bitcoin.it/wiki/MtGox/API/HTTP/v0#0.2FgetDepth.php
and also for full market depth: https://en.bitcoin.it/wiki/MtGox/API/HTTP/v1#Full_Depth
Full market depth is limited in use in time, also it is not very practical to process so distant orders data every time.
Looking at the recent trading volumes it is clear that amount of orders returned by truncated market depth api is very small. Currently it returns around 6000 btc. Comparing to daily volume of all btc traded on that market it is definitely not enough.
Is there any way to get market depth 2-5 times depper than current truncated? and do not download full depth...
As i read in unofficial api doc here: https://bitbucket.org/nitrous/mtgox-api/overview#markdown-header-moneydepthfetch
can filter_min_price and filter_max_price be an arguments for this api call?
Thx in advance
mtgox already provide api for truncated market depth: https://en.bitcoin.it/wiki/MtGox/API/HTTP/v0#0.2FgetDepth.php
and also for full market depth: https://en.bitcoin.it/wiki/MtGox/API/HTTP/v1#Full_Depth
Full market depth is limited in use in time, also it is not very practical to process so distant orders data every time.
Looking at the recent trading volumes it is clear that amount of orders returned by truncated market depth api is very small. Currently it returns around 6000 btc. Comparing to daily volume of all btc traded on that market it is definitely not enough.
Is there any way to get market depth 2-5 times depper than current truncated? and do not download full depth...
As i read in unofficial api doc here: https://bitbucket.org/nitrous/mtgox-api/overview#markdown-header-moneydepthfetch
Quote
There is probably an argument that can be used to limit the amount of data returned, however it is currently unknown
Anybody knows anything about it? is it possible to use this argument for full depth to download only small part of full depth?can filter_min_price and filter_max_price be an arguments for this api call?
Thx in advance
Jump to: