USSR Team

Вернуться   USSR Team Forum > USSR Team Projects > WRecord

WRecord NFS Undercover track records, located at WRecord.com

Ответ
 
Опции темы Опции просмотра
Старый 16.11.2009, 01:10   #51
rcgldr
Junior Member
 
Регистрация: 12.11.2009
Сообщения: 14
По умолчанию

Although my lap times aren't close to world records, I can offer some tips:

For high speed tracks like Spa and Willow Springs, the Zonda R with zero downforce or 0 front 1 rear seems to be best. My best times at Nordschleife are also done with 0/1 downforce, but I can only run about 6:15, so I'm not competitive. At Willow, I'm close at 1:08.2, but at Spa I'm only 2:04.6. For the slower tracks, the Z06 with high downforce seems to be best.

The amount of bouncing and grip in NFS Shift seems to vary depending on your PC, so you may need to adjust them depending on your system. The patch is supposed to fix the bouncing, but there is still an issue related to the video card you use (see below).

On my system, manual shift, all assists and autoclutch off is fastest. Autoclutch off with manual shift in to 1st gear before start results in very fast launch. You can shift without clutch and autoclutch off on most cars (some will spin the clutch). With ABS off, manually downshifting while braking is faster. You can also downshift into a low gear while at high speed and full throttle just before braking. This isn't needed on the Zonda R or Z06, but it helps with the EVO IX that most players use for a 10.0 car.

Some players report that setting TC to high helps. It does keep some cars, like the Z06 from bouncing as much, but I don't get any improvement in lap times, and it messes up the otherwise very fast autoclutch off launch. Other players report setting steering assist to low helps, but again, it didn't help on my system, and lap times where about the same, and I had to fight the steering assist when trying to go inside of the track when going over jumps.

Braking stops or reduces the car bouncing. At Nordschleife, there are a few problem bouncy spots where I approach fast and brake hard while going over them to prevent or reduce bouncing. However there are also problem bouncy spots where the car is accelerating, and all I can do is ease off the throttle a bit while the car bounces a little. Zonda R isn't too bad, but the Z06 is slowed down too much by this on my system.

There's also a bug that I call the "seam" bounce. Note that the surface of the track you see is not the surface of the track used for the game's physics. The game implements the physics of tracks as a huge number of flat triangles (if you've ever used the track editor for NFS4 - High Stakes, you'd know what these look like). This is aggravated when a car's suspension is compressed from a dip or on a banked track, like Rustle Creek or Dakota Tri-oval. The body of a car will snag on the seams between sets of triangles, causing the car to bounce badly and going off into the wall. The Nissan GTR SpecV (R35) used for 14.0 events will do this badly if you have a body kit installed at Rustle Creek. In my case I removed the body kit and used a seam welded chassis, it still bounces, but not as badly.

You can download this savefile which has cars setup for online play for baseline setup, except I change downforce on Zonda R to 0 front, 1 rear.

http://rcgldr.net/nfss/nfss.zip

The Zonda R in that save should be used for most tracks, and the Z06 with high downforce 44/50 for short tracks like Alpental or Tokyo. The Z06 will bottom out at about 160 mph => 255 kph, but this isn't an issue for the short tracks.

Unfortunately, the patch may not fix the stutter / peformance issue. According to this review of patch 1.2 and updated ATI drivers. From page 7 of this review, they note that the ATI HD4870X2 results in stutters and control issues compared to the HD5870, even though frame rates are the same.

Here, we see the Radeon HD 5870 and HD 4870 X2 almost as equals, with the HD 4870 X2 taking a small lead. Though the difference in framerates is small, the difference in gameplay was huge. Using the HD 4870 X2 at 2560x1600 with 24X CFAA, the game was not smooth at all. It was choppy and there was a huge amount of input latency. It took longer for the car or brake to turn after we pressed the button telling it so, and so we crashed a lot more.

http://www.hardocp.com/article/2009/...performance_iq

What this means is that the game doesn't play the same on all PC's. Not as extreme as the HD4870X2 at max settings, but there's still a difference. It's a shame because the older NFS games never had this problem. NFS2 -> NFS5 had a controller input based replay system, which guaranteed that all PC's played the same (or else the replays wouldn't work when players traded them). In 4 years or so of game play with NFS2 -> NFS5, the community of players only encountered one PC that affected the peformance of NFS4 - High Stakes (making it run 3% faster), and only when that system was overclocked; if that system was run in standard mode, it ran OK. EA took out the replay system with NFS6 - Hot Pursuit 2. I don't know if there were system sensitive issues with NFS6. NFS7 - Underground 1 - was the first NFS game known to have system sensitive issues. The original version (what you get just after installing the game) ran incredibly fast on fast systems, (but so did the game clock). The later patches reversed the situation, slower PC's got faster lap times, mostly because slower PC's didn't lose speed after landing from jumps, bouncing off walls, or just taking corners, as much as faster PC's did, depending on the track, it was about a 2% to 4% difference.

Последний раз редактировалось rcgldr, 16.11.2009 в 05:36
rcgldr вне форума   Ответить с цитированием
Старый 16.11.2009, 21:50   #52
USSRxV_NMad
Дед маздай без зайцев
 
Аватар для USSRxV_NMad
 
Регистрация: 24.03.2005
Адрес: Ukraine, Odessa
Сообщения: 12,154
Отправить сообщение для  USSRxV_NMad с помощью ICQ
По умолчанию

about 1st part of ur post:
I have only one explanations is a difference in use of control the car, each player has the personal style of driving, you can brake before turn reducing transfers, or to brake using a brake, you can brake in the turn and then open an accelerator, and can brake before turn and on a half-open accelerator to pass the turn, all these variations of control yield different results, I have noticed it on example SPA
about slow\fast PC, I agree with U 100% in NFSU\NFSU2 it was

now in the Shift If I have doubt I look telemetry and I understand where have lost or have won. Also to me it seems that in new NFS there is no dependence of configuration PC and result on a track, cuz, imho, it is the new engine for game taken from rfactor
USSRxV_NMad вне форума   Ответить с цитированием
Старый 17.11.2009, 00:13   #53
rcgldr
Junior Member
 
Регистрация: 12.11.2009
Сообщения: 14
По умолчанию

Цитата:
Сообщение от USSRxV_NMad Посмотреть сообщение
I agree with U 100% in NFSU\NFSU2
Quite a few players were aware of the issue with NFSU1, but I don't recall any discussions about PC affecting peformance of NFSU2, perhaps because the NFSU1 difference were more extreme, somtimes a differece of 10 mph => 16 kph after a drop landing (sometimes no loss in speed on slower PC's) or wall bounce on a track like Stadium. If NFSU2 had differences, it was more subtle than the issues in NFSU1, but I wasn't paying attention to lap times as much since there were so many tracks, other than Outer Ring 2 and 3 lap times, and some Bayview URL tracks where bouncing off walls wasn't used.

Цитата:
now in the Shift If I have doubt I look telemetry and I understand where have lost or have won. Also to me it seems that in new NFS there is no dependence of configuration PC and result on a track, cuz, imho, it is the new engine for game taken from rfactor
The article linked to above mentioned the difference in gameplay was huge issue with the HD4870X2 (versus the HD5870) when maxed out. Some sort of stutter issue even though frame rate was high, but perhaps that's an unusual case.

What also brought NFS Shift system sensitivity questions are the youtube videos made by ardathrokok. They are not only fast, but seem much smoother than other videos I've seen or my own gameplay (especially in the Z06). His videos look more like gameplay on the PS3 (smoother) than the PC. It is possible that the game play is smoother because he has some assists turned on: steering assist low, TC high. I tried this and it is a bit smoother, but lap times weren't affected significantly.

standing start 6:05.690 in Z06:
http://www.youtube.com/watch?v=aNaEPUnxs68&fmt=18

standing start 6:45.450 in works Audi TT:
http://www.youtube.com/watch?v=cQ8BCwj5CWI&fmt=18

I was thinking a good setup could cause this, but these were done with default setups according to his thread here:

http://forum.nfsunlimited.net/viewto...?f=140&t=17940

I'd be curious as to what other players can run at Nordschleife with this configuration, call it the "ardathrokok challenge", works Z06, default setup, perhaps with the same assists: steering assist low, traction control high, all other assists off. With this combination of assists, you can usually go full throttle mid turn and the corner exit won't end up with oversteer.

Последний раз редактировалось rcgldr, 17.11.2009 в 05:08
rcgldr вне форума   Ответить с цитированием
Ответ


Здесь присутствуют: 1 (пользователей - 0 , гостей - 1)
 
Опции темы
Опции просмотра

Ваши права в разделе
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is Вкл.
Смайлы Вкл.
[IMG] код Вкл.
HTML код Выкл.

Быстрый переход


Часовой пояс GMT +1, время: 14:25.


Powered by vBulletin Version 3.8.12 by vBS
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.