#4419022 - 05/03/18 10:58 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: UnderTheRadar]
|
Joined: Jun 2005
Posts: 16,082
- Ice
Veteran
|
Veteran
Joined: Jun 2005
Posts: 16,082
Philippines / North East UK
|
If they tested on robust machines they may not have seen the issue. I had 16 Gigs in my system and had no problems. And does proper testing not include testing your product on less robust machines to find out what your minimum spec requirements are? Are there no DCS players who play on laptops? I think this would've been fine if ED started out by saying "Um, guys, 2.5 is more resource-hungry and we may need more RAM to run it properly!" and that would show ED knows what they're doing. Instead, we see "Whoops!! Our program doesn't run well with less than 8GB RAM and our customers have noticed! Let's update the minimum specs page!"This is also ignoring the fact that many on the ED forums have stated that 2.5 should run close to Normandy performance levels. Were we clamoring for 16GB and 32GB RAM kits to run Normandy?
- Ice
|
|
#4419062 - 05/04/18 08:11 AM
Re: 2..5 Release Date ANNOUNCED!!
[Re: - Ice]
|
Joined: Mar 2003
Posts: 3,922
Paradaz
Senior Member
|
Senior Member
Joined: Mar 2003
Posts: 3,922
UK
|
You're countering your own argument there........You say you're sure they know about the problems before release, but precluded that with a sentence about QA testing with minimum spec machines. Given that the minimum spec is now 8GB for low settings (and was probably 4GB before the spec alteration?)......a multiplayer test would have highlighted the memory problem straight away. As far as I can see there are only three possible explanations....
1) ED don't test properly 2) They didn't know about the memory problem (relates directly to point 1) 3) They knew about the memory problem and didn't bother telling anyone but put the software in open beta anyway
Given their statements and forum activity, I'd suggest it's 1 and 2, but some may argue it's 1 and 3. Either way and regardless of how anyone looks at it, ED have just confirmed yet again they are incompetent and haven't got a clue how to efficiently integrate, test and release.
As for your statement....you don't 'hope' in software development, you test, verify and understand the problems due to the effects on resources and funding that any issues will have. There will always be levels of risk any software development, however in this case there is no excuse...you don't 'hope' when you can completely eliminate the risk via very simple testing. Although on the same token you do have a point......because ED haven't got a clue what they're doing, they probably do 'hope' and that's exactly why they can never align to timelines and subsequently undoubtedly why they hemorrhage funds and ultimately their own profit margins because they have no control over how long their development activities are taking.
On the Eighth day God created Paratroopers and the Devil stood to attention.
|
|
#4419126 - 05/04/18 04:45 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: Slippery_Rat]
|
Joined: Jun 2005
Posts: 16,082
- Ice
Veteran
|
Veteran
Joined: Jun 2005
Posts: 16,082
Philippines / North East UK
|
Ah, I see we've moved on yet again; previous points abandoned but can't be bothered to concede gracefully after fighting dirty? Any proper QA/QC department Yes. A **proper** QA/QC department is what ED does not have. I'm sure they know the problem before release. It's already been demonstrated that this is not the case. It's not beyond all reason to update the minimum specs at some point. Again, comparing ED with any other decent company is where you're going wrong. It's fair enough to update min specs, but that's not really what ED has done now, was it? Let's not go through any mental gymnastics here. You do not warn people of higher minimum specs **AFTER** the community figures it out for themselves. 2) They didn't know about the memory problem (relates directly to point 1) This. If they knew it was a memory allocation problem, then that would've been the story straight away, instead of calling it a memory leak then changing it to a memory allocation issue. While not reading all details of this argument, I suspect the problem is video memory not system memory Nope, increasing the recommended specs from 4GB to 8GB or 8GB to 16GB clearly demonstrates is system memory and not GPU memory.
- Ice
|
|
#4419131 - 05/04/18 05:03 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: Slippery_Rat]
|
Joined: Jun 2002
Posts: 850
KeyCat
Member
|
Member
Joined: Jun 2002
Posts: 850
Sweden
|
While not reading all details of this argument, I suspect the problem is video memory not system memory, after all the programming code probably hasn't changed that much but what has changed, is the resources required to draw to the screen.
Again, I'm blind here, not having tried it myself.
As other said I don't believe this is the only issue. Looking at the DCS forum there are reports of crashes left to right as well as AI, MP, lighting issues and many other issues. I'm quite surprised of current state of DCS since they worked on the new EDGE graphics engine for ~5 years already. Also wonder how all this wasn't discovered by external beta testers, I mean they must be more than two guys (+ ED's internal testers)? One thing for sure, I don't envy them and I'm glad I'm not in their situation (as a small business owner myself). I really like the DCS concept but not in the state it's been the past years. I'm sure they have good intentions and I wish ED all the best and hope they will get their "vision" to work as expected before I'm 6 feet under Just my 0.20 SEK and that's not worth much, ask theOden - LOL!
>> It's all about teamwork! <<
|
|
#4419139 - 05/04/18 05:23 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: KeyCat]
|
Joined: Jun 2005
Posts: 16,082
- Ice
Veteran
|
Veteran
Joined: Jun 2005
Posts: 16,082
Philippines / North East UK
|
Also wonder how all this wasn't discovered by external beta testers, I mean they must be more than two guys (+ ED's internal testers)? Maybe because there aren't any? Some would argue that ED does have testers but with the things that made it past those guys, I wonder what they really do as "testing"?
- Ice
|
|
#4419159 - 05/04/18 06:48 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: - Ice]
|
Joined: Sep 2014
Posts: 797
leaf_on_the_wind
Member
|
Member
Joined: Sep 2014
Posts: 797
|
My 2c
It is not a memory leak in the conventional sense, it is piss poor memory management
Before DCS World went to 64bit only, it was restricted to max of 3G of RAM and it worked fine back then Since they moved to 64bit, the RAM usage has increased a lot, from what I have tried pre-load radius at 75 = 6G of RAM pre-load radius at 150 = 12G of RAM used (withrout even doing much else - just loading the game up and starting)
They need to optimise how much RAM is used and optimise the game engine itself, throwing more resources eg RAM at it is NOT the solution My guess ...... speed tree is why this is happening now
Ferengi Rule of acquisition #1 Once you have their money ... never give it back.
|
|
#4419165 - 05/04/18 07:53 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: leaf_on_the_wind]
|
Joined: Mar 2011
Posts: 2,042
cichlidfan
Member
|
Member
Joined: Mar 2011
Posts: 2,042
Woodbridge, VA, USA
|
My guess ...... speed tree is why this is happening now
I suspect your guess is incorrect since Speedtree is not used on the Normandy or Nevada maps.
ASUS ROG Maximus VIII Hero, i7-6700K, Noctua NH-D14 Cooler, Crucial 32GB DDR4 2133, Samsung 950 Pro NVMe 256GB, Samsung EVO 250GB & 500GB SSD, 2TB Caviar Black, Zotac GTX 1080 AMP! Extreme 8GB, Corsair HX1000i, Phillips BDM4065UC 40" 4k monitor, VX2258 TouchScreen, TIR 5 w/ProClip, TM Warthog, VKB Gladiator Pro, Saitek X56, et. al., MFG Crosswind Pedals #1199, VolairSim Pit, Rift CV1
|
|
#4419229 - 05/05/18 07:47 AM
Re: 2..5 Release Date ANNOUNCED!!
[Re: KeyCat]
|
Joined: Aug 2010
Posts: 1,614
theOden
Member
|
Member
Joined: Aug 2010
Posts: 1,614
|
Just my 0.20 SEK and that's not worth much, ask theOden - LOL!
Well, SEK has a unique awesomeness to it so I personally value it more than EUR. Anyhow, I just fired up DCS this morning to have a look at the miserable state of affairs. Most of my SP missions use something called "world.searchObjects" to find and list bridges, factories, warehouses etc. but this function is now caputt rendering all my missions useless (it does still find static object types though, but yeah). Seriously, how on earth can you #%&*$# something up this much? I know Han (he who left for new IL2 series) have some rather hard words for ED but I'm starting to think he frigging correct on each account (incl. his view on ED community). Time to let the blue DCS icon rest a few weeks again.
|
|
#4419251 - 05/05/18 06:08 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: leaf_on_the_wind]
|
Joined: Jun 2005
Posts: 16,082
- Ice
Veteran
|
Veteran
Joined: Jun 2005
Posts: 16,082
Philippines / North East UK
|
I know Han (he who left for new IL2 series) have some rather hard words for ED but I'm starting to think he frigging correct on each account (incl. his view on ED community). Link? It is not a memory leak in the conventional sense, it is piss poor memory management Show me one other game where a vdieo memory manager is a feature announced in the newsletter? It's like advertising your game works with a keyboard and mouse and runs on 1080p!!
- Ice
|
|
#4419271 - 05/05/18 08:43 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: - Ice]
|
Joined: Aug 2010
Posts: 1,614
theOden
Member
|
Member
Joined: Aug 2010
Posts: 1,614
|
Was referencing this: http://SimHQ.com/forum/ubbthreads.php/topics/4410535/head-of-il-2-han-posts-about-dcs#Post4410535But since google translate is needed I honestly think I read it wrong first time, maybe influenced by Sobek's comment there. Contrary to my first impression, reading it now I think he's almost defending ED World turns and twists. Regarding memory usage, yes something went tits up but please all - try to remember that not using available memory is really not a sign of good memory management. What I mean is, while running the app SHOULD use all available memory. Right?
|
|
#4419380 - 05/06/18 06:10 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: theOden]
|
Joined: Jun 2005
Posts: 16,082
- Ice
Veteran
|
Veteran
Joined: Jun 2005
Posts: 16,082
Philippines / North East UK
|
Regarding memory usage, yes something went tits up but please all - try to remember that not using available memory is really not a sign of good memory management. What I mean is, while running the app SHOULD use all available memory. Right? Yes but a good program should be able to adjust whether the user has 4GB, 8GB, 16GB, or 32GB RAM and use only what is needed. A program should not use 30GB RAM when it only needs 4GB and a program that needs more space should be able to use the RAM properly and even the pagefile. Deteriorating performance as the program hits RAM limits is a coding issue.
- Ice
|
|
#4422392 - 05/24/18 08:17 AM
Re: 2..5 Release Date ANNOUNCED!!
[Re: - Ice]
|
Joined: Apr 2014
Posts: 871
Winfield
model citizen
|
model citizen
Member
Joined: Apr 2014
Posts: 871
QLD
|
I thought there would be some updates to this after yesterday's update.
DCS 2.5.2.17559 in the Open Beta
The Persian Gulf map is available.
DCS World Su-25T: training mission CN voiceover fixed In-cockpit mirrors will not have red outline. Sound - reworked guns sources loading routine to allow aircraft-module-specific samples to be used. Restored texture of KDO G40
DCS A-10C A-10C: training mission CN voiceover fixed
DCS Flaming Cliffs FC3: added A-10A CN flight manual
DCS MiG-21Bis by Magnitude 3 LLC Created new indicator lights. Fixed textures on English ASP switches and other objects. Adjusted red flood-lights to be brighter. Adjusted pink color lights to red-ish Fixed see-through areas in cockpit. Fixed flap and ARK buttons' position animation. Corrected SPO's center area not to rotate when switching between night/day modes. Fixed RSBN NPP Glideslope animations. Fixed RPM 2 needle not overtaking RPM 1 needle. Fixed non-rotating compressor blades. Fixed landing gear strut lights from being on constantly. Fixed a bug that caused small errors in aerodynamic forces calculation for the landing gear. Slight improvements in "Game Flight Mode" and/or "Simplified Engine Management" is on. Engine won't shut down on high lateral or high negative longitudinal acceleration (such are in spins) Engine won't stop at low oil pressure time trigger. Engine won't shut down at low fuel pressure. Added Iran, Iraq and Syria liveries by Jack
DCS AV-8B N/A by Razbam Simulations LLC IR Maverick slew and lock updates Stabilator indicator correction Corrected water flowing on and off rapidly in Landing STO mode Corrected MPCD glow not going off when MPCD goes off after being on Corrected minor shake due to gear model when not moving with parking brake Increase canopy open/close speed to match video references Canopy open/close keyboard shortcut now automatically performs unlock handle Corrected Radio COMM 2 not working in UFC mode Corrected ARBS/DMT TDC bug. IR/TV Mavs launch constrain keyhole is now active. AGM-65Es are now active. MPCD behavior after firing a Maverick has been updated. IR/TV Mavs axis slew is now active. STRS Page WPN FAIL message bug is now fixed. (The bug appears when using mavericks). Weapon AUTO (CCRP) release mode improved logic, cues, and accuracy Weapon CIP (CCIP) release mode is upgraded. Still WiP. NVG Brightness control
DCS M-2000C by Razbam Simulations LLC Vortices/vapor effects Corrected issue when on engine idle on ground and subsystems not receiving power correctly Corrected endless refueling when on ground and requesting less internal fuel than currently in aircraft Corrected AI main gear pos contact point slightly above ground Corrected linearity in pitch control for consistent pitch down response
DCS SA342 Gazelle by Polychop Simulations Gazelle TV black screen fixed
DCS UH-1H by BST UH-1H: Corrected "Rebel Base" mission (by slpak). Removed forest in target zone.
MiG-15bis by BST Fixed game freezes when ADF switch is enabled first time
DCS AJS-37 Viggen by Heatblur Simulations Fix to memory corruption issue causing inoperable HUD/ADI and other systems Fix to issue where flight plan speed would be ignored if force speed was set Fixes to auto trim and pitch stabilization Corrected Kneeboard, map is now visible Added missing image to NAV - tutorial mission Fix to TILS system when flying fast Corrections to rocket CCIP sight accuracy.
Campaigns F-5E ACM Campaign - corrected missions with F-4E. F-5E BFM Campaign - corrected missions with F-4E. F-15C Aggressors BFM Campaign - corrected missions with F-4E. A-10C BFT Campaign - mission BFT05 corrected. L-39 Kursant Campaign - missions 2, 4, 5, 7 corrected. A-10C Advanced Aircraft Training Campaign - corrected altitude triggers, updated some dialogue text, added mission success music.
Known issues Player parkings on carrier are broken. External lights (flashlight, taxi light) are visible inside some cockpits. Spitfire tail wheel contact model with ground is broken.
|
|
#4430320 - 07/14/18 02:48 PM
Re: 2..5 Release Date ANNOUNCED!!
[Re: - Ice]
|
Joined: Nov 2001
Posts: 3,955
Sokol1
Senior Member
|
Senior Member
Joined: Nov 2001
Posts: 3,955
Internet
|
Soon™
DCS World Dedicated Server Update
With the launch of 2.5 earlier this year, we have been working diligently on an item we've been wanting to bring to DCS World for quite a long time: a dedicated server. It however was not until the launch of DCS World 2.5 was this possible, and hence the length of time this has taken. As a dedicated server, the software will not require any graphics rendering, so no dedicated graphics card will be required and there is no audio requirement. This all greatly reduced server load overhead. There will be a new web GUI to control the server. Operation of the dedicated server through the web GUI will operate much as the existing online server does now, but allow you to make changes remotely. We plan to release the dedicated server in Autumn 2018. Afterward, we will investigate adding additional features to expand its capability.
|
|
|
CD WOFF
by Britisheh. 03/28/24 08:05 PM
|
|
|
|
|
|
|
|
|
|
|