表示
请求上传的client数由CKnownFile中的列表决定
Rev61 Enig123
code cleanups
2007.4.9
Rev60 Enig123
Fixed a bug that handle lowid callback in a wrong way. Stupid me!
Rev59 Enig123
Several minor fixes: only count randCS itself when the process is only related to the single client
2007.4.8
Rev58 Enig123
Gracefully handle lowid clients that was marked to upload and still waiting. Count it's randCS increase rate proportional to the ratio it should get AFTER it get it's share.
Rev57 Enig123
Let lowid client's randCS increase when waiting and marked as been chosen again
Rev56 Enig123
Lovelace Credit change: Client download from me more than upload to me + 1/16 bonus, it get credit ceiling of 20
Rev55 Enig123
Give some room for extra upload due to big randCS
Rev54 Enig123
Add extra log message for more than one part due to positive randCS
Rev53 Enig123
Fixed bug with upload more than one part with directly selected uploading client
Rev52 Enig123
Fixed a bug compiling with CryptoPP 5.4
Rev51 Enig123
Give more than one chunk if the upload client was been chosen directly (aka. randCS > PARTSIZE)
2007.4.6
Rev46 Enig123
Credits Reset Exploit Prevention from NetFinity's Warp
Rev45 Enig123
Several fixes and improvement from Xtreme & Spike2 mod
2007.4.5
Rev44 Enig123
Code cleanup: remove unused code
Rev43 Enig123
minor fixes and cleanups
2007.4.4
Rev42 Enig123
UploadQueue minor change
BaseClient minor parameter change
2007.4.3
Rev41 Enig123
Smooth global download speed graph
Rev40 Enig123
As Xman did, 80% for non SUI client
Rev39 Enig123
GetScoreCN definition change
Take file priority into account when computing Scores for randQueue selection
Rev38 Enig123
PNG invoke redesigned (formerly as a member of class)
Rev37 Enig123
Some more optimizations from EMF 0.3
2007.4.2
Rev36 Enig123
Added files for Kademlia helper
Rev35 Enig123
KAD thread proof (aka. safe kad + kad helper) from EMF 0.3
Rev34 Enig123
Optimization of routine RemoveSource() in DownloadQueue.cpp from EMF 0.3
Rev33 Enig123
Some performance optimizations from EMF 0.3
Rev32 Enig123
Safe Tag Sending optimization from EMF 0.3 within method of ::SendMuleInfoPacket()
Rev31 Enig123
Fix a bug with randQueue selection method, which may cause trouble in rare case
Rev30 Enig123
Parameters adjust for speed calculation
Add extra 0 item for single client speed calculation for better time resolution
2007.3.31
Rev29 Enig123
Global download speed calculation minor fix: no more extra 0 item fills.
Rev28 Enig123
Small client upload speed calculation fix (better display now)
Rev27 Enig123
Fixed bug of global download speed calculation brought by last rev
Rev26 Enig123
Download Speed Calculation Fix
2007.3.30
Rev25 Enig123
don't give randCS to lowid client that already been chosen
2007.3.29
Rev24 Enig123
Unnecessary 0 value refresh on upload speed calculation
Rev23 Enig123
Global Download Speed Calculation also uses ACATSpeedMeasurer
Some other fixes for speed calculation
Rev22 Enig123
Proper SpeedMeasurer initialization
Rev21 Enig123
Global Upload speed calculation cleanups
Rev20 Enig123
UInt128.cpp minor improvement from NetFinity
Rev19 Enig123
ListenSocket.cpp reverted
2007.3.28
Rev18 Enig123
Download speed calculation fix
Rev17 Enig123
Speed code cleanups
Rev16 Enig123
ACATSpeedMeasurer refinement
Rev15 Enig123
handle special case of ACATSpeedMeasurer
2007.3.27
Rev14 Enig123
ACATSpeedMeasurer minor fix
Rev13 Enig123
More accurate speed calculation.
Download speed measure bug fix.
2007.3.26
Rev12 Enig123
Disable AICH request response personally.
Rev11 Enig123
Credits calculation minor modification.
Rev10 Enig123
ACATSpeedMeasurer Reinvented, preparing for further 2hr speed calculation.
2007.3.25
Rev9 Enig123
Don't decrease randCS value if there's no other clients within waiting queue to be compensate
Rev8 Enig123
TitleMenu fix
2007.3.24
Rev7 Enig123
Random Selection start at random position instead of fixed 0
Rev6 Enig123
correctly set CreditSystemMode as enum. Also treat UploadChooseAlgorithm as enum
move definition of CreditSystemMode to Preferences.h
Rev5 Enig123
Upload Choose Algorithm as enum
Rev4 Enig123
randCS Display added
踩姑娘的小蘑菇@2007-04-05 17:56
恩,等待放出~
fatwaipersonal@2007-04-05 18:15
速度計算系統重寫,應該可以降低資源占用(特別是內存)
這個很重要,近來運行了一會後便佔用了我400MB內存
闇夜@2007-04-05 18:22
引用
最初由 fatwaipersonal 发布
速度計算系統重寫,應該可以降低資源占用(特別是內存)
這個很重要,近來運行了一會後便佔用了我400MB內存
400M
太[/han]了
openicq@2007-04-05 19:32
oh 支持~
我EM内存使用一般不超过30M 虚拟内存不超过50M
thwj@2007-04-05 19:53
等着新版本试用了,现在的em我还是觉得占资源太多了,弄不弄50M以上……
Lwy@2007-04-05 20:12
上传算法有自选么,总是觉得很诡异
zhouwei_e@2007-04-05 20:18
算法当然可以选择
fatwaipersonal@2007-04-05 21:05
引用
最初由 闇夜 发布
400M
太[/han]了
不知是否排隊列表中人多,或是我開上傳太高才會這樣:confused:
PS:上傳設為2M/s,上傳列表中大約200人,排隊列表中大約200~300人
zhangshaox@2007-04-05 21:58
驴榜证书功能加回来没有?
tengfy@2007-04-05 22:21
学校用不了EM,只能等回家再试看了--
yanzi@2007-04-05 22:33
等待中.....
内寸要400M是有点大
lvtianhui@2007-04-05 22:53
我都是默认的 用的VC版的 为什么占到120m 120m呢
速度也只有120 120啊
zhouwei_e@2007-04-05 23:37
引用
最初由 zhangshaox 发布
驴榜证书功能加回来没有?
貌似就没去掉吧?
zhouwei_e@2007-04-05 23:43
引用
最初由 yanzi 发布
等待中.....
内寸要400M是有点大
400M这个应该是内存泄漏了,直接换版本吧
3楼用的什么版本,貌似CN版不至于这样吧
«123»共3页
| TOP