Eadeqa
|
|
May 01, 2014, 08:16:22 AM |
|
Looking forward to test the new version! An advice from an old c++ coder: bugfix: double-round function was not accurate enough for one comparision
Don't use floating point in a financial application! Nasty round and compare bugs ahead! Use a long or longlong and do integer division to get to the decimals. Yes that might be a good idea;-) Right now it will round it to 12 decimals accuracy and is working as well.... but we will consider changing that. Make that priority, please. Before the release date. Floating points should not be used in financial applications.
|
|
|
|
|
|
|
|
TalkImg was created especially for hosting images on bitcointalk.org: try it next time you want to post an image
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
|
|
ThomasZ80
Newbie
Offline
Activity: 24
Merit: 0
|
|
May 01, 2014, 09:42:28 AM Last edit: May 01, 2014, 09:53:09 AM by ThomasZ80 |
|
... sent 3333 EXO right back!!! Got it! EXOk3Z2aewR5asfdpT6qCFcO5gxDtTNDyw4tdbhg
|
|
|
|
eXo_coin (OP)
|
|
May 01, 2014, 10:05:22 AM |
|
... sent 3333 EXO right back!!! Got it! EXOk3Z2aewR5asfdpT6qCFcO5gxDtTNDyw4tdbhg network looks good right now ... lets wait some more hours! I was very generous transfering 0.1 EXO
|
|
|
|
wizzardTim
Legendary
Offline
Activity: 1708
Merit: 1000
Reality is stranger than fiction
|
|
May 01, 2014, 10:10:22 AM |
|
PLease send me some for test EXO6ZqUDk8rexbs4uPztfK69WKUYPATSSrRndbgm
|
Behold the Tangle Mysteries! Dare to know It's truth.
- Excerpt from the IOTA Sacred Texts Vol. I
|
|
|
wizzardTim
Legendary
Offline
Activity: 1708
Merit: 1000
Reality is stranger than fiction
|
|
May 01, 2014, 10:11:24 AM |
|
... sent 3333 EXO right back!!! Got it! EXOk3Z2aewR5asfdpT6qCFcO5gxDtTNDyw4tdbhg sent you some
|
Behold the Tangle Mysteries! Dare to know It's truth.
- Excerpt from the IOTA Sacred Texts Vol. I
|
|
|
Fonzarelli
Newbie
Offline
Activity: 17
Merit: 0
|
|
May 01, 2014, 10:12:48 AM |
|
I just was messing around with 666.666 EXO However, client sent 666.667 So I created EXO out of nowhere
|
|
|
|
ThomasZ80
Newbie
Offline
Activity: 24
Merit: 0
|
|
May 01, 2014, 10:12:55 AM |
|
I was very generous transfering 0.1 EXO WOW! Thanks a lot. Got it.
|
|
|
|
eXo_coin (OP)
|
|
May 01, 2014, 10:14:14 AM |
|
PLease send me some for test EXO6ZqUDk8rexbs4uPztfK69WKUYPATSSrRndbgm
send: TX55935C7CA5900A81108DF73336EDA8E363A65D8AC35B46D29428F8DC17E6954C btw: mining will hopefully start at block 100 automatically (watch block explorer mining value or mining tab)
|
|
|
|
wizzardTim
Legendary
Offline
Activity: 1708
Merit: 1000
Reality is stranger than fiction
|
|
May 01, 2014, 10:14:23 AM |
|
|
Behold the Tangle Mysteries! Dare to know It's truth.
- Excerpt from the IOTA Sacred Texts Vol. I
|
|
|
ThomasZ80
Newbie
Offline
Activity: 24
Merit: 0
|
|
May 01, 2014, 10:15:22 AM |
|
... sent 3333 EXO right back!!! Got it! EXOk3Z2aewR5asfdpT6qCFcO5gxDtTNDyw4tdbhg sent you some Just got 333.33.
|
|
|
|
Fonzarelli
Newbie
Offline
Activity: 17
Merit: 0
|
|
May 01, 2014, 10:19:17 AM |
|
Does the bug reply section already work?
|
|
|
|
eXo_coin (OP)
|
|
May 01, 2014, 10:23:39 AM |
|
I just was messing around with 666.666 EXO However, client sent 666.667 So I created EXO out of nowhere Yup... got it! The eXominer seems to be working too... but I have a question... why does it stop hashing away in order to submit to parent? that is the key innovation regarding our mining system: if you find a system you submit it and therafter up to the start of the next block your miner will idle. ->Saving you energy costs! The faster you find a result the more EXO you'll get and the less energy cost you will have (procentually regarding hashpower). As stated, you'll get hopefully some mining transactions after block 100 has been reached
|
|
|
|
timmyd
|
|
May 01, 2014, 10:37:09 AM |
|
Im liking what im reading here
|
|
|
|
mikesbmw
|
|
May 01, 2014, 10:38:38 AM |
|
I just was messing around with 666.666 EXO However, client sent 666.667 So I created EXO out of nowhere What's causing this? Is this some floating point bug? Do you use floating point at all?
|
|
|
|
chinnuperiya
Newbie
Offline
Activity: 34
Merit: 0
|
|
May 01, 2014, 10:43:40 AM |
|
Looking forward to test the new version! An advice from an old c++ coder: bugfix: double-round function was not accurate enough for one comparision
Don't use floating point in a financial application! Nasty round and compare bugs ahead! Use a long or longlong and do integer division to get to the decimals. Yes that might be a good idea;-) Right now it will round it to 12 decimals accuracy and is working as well.... but we will consider changing that. Make that priority, please. Before the release date. Floating points should not be used in financial applications. yes. I have had nightmares working with double in financial modelling. Please change it to long before the final release
|
|
|
|
Fonzarelli
Newbie
Offline
Activity: 17
Merit: 0
|
|
May 01, 2014, 10:45:26 AM |
|
I just was messing around with 666.666 EXO However, client sent 666.667 So I created EXO out of nowhere What's causing this? Is this some floating point bug? Do you use floating point at all? I sent about 666.666666666666666666 in the form Exo-client sent 666.667 so he round to 3 decimals. I dont know how is that so. But actually 666.667 left the wallet. So no "problem" at all, just rounding issue
|
|
|
|
ATC
Newbie
Offline
Activity: 49
Merit: 0
|
|
May 01, 2014, 10:49:49 AM |
|
Looking forward to test the new version! An advice from an old c++ coder: bugfix: double-round function was not accurate enough for one comparision
Don't use floating point in a financial application! Nasty round and compare bugs ahead! Use a long or longlong and do integer division to get to the decimals. Yes that might be a good idea;-) Right now it will round it to 12 decimals accuracy and is working as well.... but we will consider changing that. Make that priority, please. Before the release date. Floating points should not be used in financial applications. +1 Make it priority! Only integer in finance. BTW, anyone with address in the recent posts will get 8888.88 exo.
|
|
|
|
eXo_coin (OP)
|
|
May 01, 2014, 10:50:42 AM |
|
hm strange.. when searching for EXORGxHC1e1h2Dq6QgNpnhy32gTbh2BC2gFycmgc in the block explorer the 666.667 does not appear. The other address to that transaction is displaying the amount. That will be investigated - as well as the rounding of 667. But that are just minor bugs. If all nodes survive the next 12 hours then we are happy It is not a problem with using doubles at all. They have a precision of 15 decimals, we capped it at 12. So the 12th decimal will be a rounded value but not the third one like here. But let us check whats the issue here. Doubles are an issue in comparisions if they are not properly addressed. But with capping/rounding at 12th decimal there should not be an issue anyway. I, personally, never got in trouble with this processing of data but correct me if I am wrong. We could use unsigned long long anyway for V0.6 Right now after ~7 hours running nodes (some ~10) all are still in sync and fully functional. That's the all-time best message here I guess ;-) No crashes or asyncs yet. Crossing fingers for the next hours... #edit I just was messing around with 666.666 EXO However, client sent 666.667 So I created EXO out of nowhere What's causing this? Is this some floating point bug? Do you use floating point at all? I sent about 666.666666666666666666 in the form Exo-client sent 666.667 so he round to 3 decimals. I dont know how is that so. But actually 666.667 left the wallet. So no "problem" at all, just rounding issue Ah ok exokk most possibly capped the counting decimals of transactions within the GUI interface to 3 decimals and rounds it properly. I'll ask him/check it and will increase it
|
|
|
|
griffinriz
|
|
May 01, 2014, 10:51:00 AM |
|
Received email
|
|
|
|
chinnuperiya
Newbie
Offline
Activity: 34
Merit: 0
|
|
May 01, 2014, 11:14:15 AM |
|
please send me some test EXOs. will send them back with simple interest of 100% EXOyahgZ5Csz31ctmnK22arfAUUWAaZRs1gxdbhd
|
|
|
|
|