Jump to content
Corsair Logo Corsair Logo
Corsair Community
CORSAIR COMMUNITY Survey, we want to hear your feedback!

Discuss Link v4.6.0.86 here

Technobeard

By Technobeard
in Corsair Link

Recommended Posts

  • Replies 187
  • Created
  • Last Reply
zorvic

zorvic

Posted
Posted

http://i.imgur.com/48alu6n.png

 

Here is the previous version i had with the ability to change the LEDs

 

http://i.imgur.com/Ov0V5zP.png

 

 

Here you can see my Link dashboard. I use to be able to change my ram LED from default to static but now that option is gone. I see there is an option now to change the ram LED to both be static but its not working now. Before I updated the software only one LED blinked.

Link to comment
Share on other sites
Breimeister

Breimeister

Posted
Posted
I decided to try Corsair Link 4.6.0.86 and I managed to get everything set up properly. Shortly thereafter every time I start up Corsair Link the Asus bios shuts down the system. When the system boots up again the bios reports that power supply surges were detected. Is this a real problem with my AX860i or a problem that Corsair Link some how creates? I do not want to disable the bios anti-surge protection feature.
Link to comment
Share on other sites
flicks09

flicks09

Posted
Posted
My sincere thanks to Corsair, with this update I receiverd 4 fans for free. At least Link is showing me 8 fans now, 1 is identical with 5, 2 with 6 and so on.

 

 

I am having this issue as well with 4.6.0.86. Fan 1 is also fan 5, fan 2 repeated as fan 6 and fan 3 repeated as fan 7.

 

I have the H110i cooler.

 

Same issue for me too. 4 new fans, 8 in total, with duplicates. Whhhhhhy.

Link to comment
Share on other sites
specialkone

specialkone

Posted
Posted
Same issue for me too. 4 new fans, 8 in total, with duplicates. Whhhhhhy.

 

I'm asking why as well. I didn't have this issue in any of the previous versions of link. I would go back to 4.5.055 but sticking with 4.6.086 because of this comment in the 4.6.086 release notes which now concerns me with using previous versions:

 

"Addressed issue with H110i that may run pump at lower RPM until software startup"

 

I have Corsair 750D case, Corsair H110i cooler, Corsair Vengeance memory, Corsair K70 Rapidfire keyboard and Corsair M65 Pro mouse. C'mom Corsair, fix this issue quick or give us the extra fans you are reporting free for being good customers.

Link to comment
Share on other sites
skizott

skizott

Posted
Posted

What I'd really like is for product management to listen to engineering/R&D/QA instead of some product management turd forcing engineering to release a product that, obviously, needs to spend more time in QA.

Sure, you have RGB RAM modules, there was no need to rush the release and, in turn, force engineering to release a version of Link that everyone, everyone, must have know had significant issues. Now, what does PR have to say? That is what I want to know, too.

Someone needs to ball-up, tell the community they've made series of mistakes due to a hierarchical system that is causing Corsair to fail. Some lunatic is forcing male bovine excrement down others' throats because he or she has no idea what they are doing and feel threatened if they look like they're the coital lunatic they are.

Link to comment
Share on other sites
PrEzi

PrEzi

Posted
Posted
What I'd really like is for product management to listen to engineering/R&D/QA instead of some product management turd forcing engineering to release a product that, obviously, needs to spend more time in QA.

Sure, you have RGB RAM modules, there was no need to rush the release and, in turn, force engineering to release a version of Link that everyone, everyone, must have know had significant issues. Now, what does PR have to say? That is what I want to know, too.

Someone needs to ball-up, tell the community they've made series of mistakes due to a hierarchical system that is causing Corsair to fail. Some lunatic is forcing male bovine excrement down others' throats because he or she has no idea what they are doing and feel threatened if they look like they're the coital lunatic they are.

 

If you would follow the Corsair Link development for a longer period of time then you would notice, that the Link had issues from the start (were reported mostly by red-ray several times and with each release he posts them anew). Some unfixed issues are dated 2014....

 

Honestly I have a feeling that red-ray fixed/helped more that the Link-Developers and each and every Corsair-employee that ever accessed this forum altogether...

 

I (and many others) won't touch Corsair products with a six meter pole thanks to this bad (software) support.

Link to comment
Share on other sites
FALE

FALE

Posted
Posted
Looking in http://forum.corsair.com/forums/showthread.php?p=896140 you will see a Radeon HD 7800 shows up and I expect that is using an older driver. Which driver version are you using?

 

I know the AMD Radeon Software Crimson Edition drivers report random and incorrect voltages as I needed to make SIV to filter them out and as I recall CL4 stopped reporting AMD GPUs about the same time. If you look there are lots of posts such as http://forum.corsair.com/forums/showthread.php?p=893342 about this.

 

Franck (CPUZ author) mentioned that CPUZ was also reporting silly AMD GPU voltages the other day.

 

I'm using 17.3.3 Crimson Relive. I havent had the card very long so I don't have the "it used to work" frame of reference. Not sure what the story is but oh well. Thanks for your research.

Link to comment
Share on other sites
volcom0903

volcom0903

Posted
Posted
well, i can see temp from my 960evo now, but the firmware update for LNP has bricked fan led control.

i can still control the light strips on channel 1, but the link on channel 2 that goes to my HD fan hub now only controls the first LED on the 2nd fan. the rest of the LED on that fan, and the other 3 fans are all stuck on static, purple, low brightness.

 

consolidating the k17 loads into one line was good. still no temp output (1700x, asus b350m-a)

 

<edit>

def was the firmware update that i received after updating CL4 that bricked the LNP. can still control all 4 strips independently with SIV as well. fans still stuck on purple.

 

 

I'm in the same boat but my fans are bricked completely and 3 fans led's do not work what so ever on any color. 1 fan has 1 LED that works but is flickering. This all occurred right after the update

Link to comment
Share on other sites
red-ray

red-ray

Posted
Posted

The sequence of ordering for the DIMMs is rather confusing :eyebuldge. It starts off OK, but as for 8 9 24 10 :rofl:

 

If this the intended ordering and if so please can you explain why this is sensible?

 

The system also has 32 DIMMs rather than the 24 CL 4.6.0.86 reported, see http://forum.corsair.com/forums/showthread.php?p=890726. Do Corsair actually test CL4 at all, I can't see how they can :confused:.

 

Displaying (tRC... rather than (tRCD) is either silly or dumb :sillygrin.

 

CL4 also fails to fit on a 1024 x 768 screen when it should do.

 

Note the temperature of 8.3°C in not a CL4 issue, but rather an issue in the BIOS or ACPI setup, however it did fail to report the motherboard/system name and the SMSC SCH4307 information :p:.

 

attachment.php?attachmentid=28468&stc=1&d=1490904669

RXD.thumb.png.8d048f0ce5a79de8ed27dbda51af1167.png

Link to comment
Share on other sites
red-ray

red-ray

Posted
Posted
My H100i v2 cooler has disappeared from CL4. After the patch install it was showing, then I used CL4 to change the LED color on the cooler and it is no longer displaying. Incidently the LED is now showing the default white again after reboot.

 

Does it show up on Devices and Printers or in Device Manager?

Link to comment
Share on other sites
skizott

skizott

Posted
Posted
If you would follow the Corsair Link development for a longer period of time then you would notice, that the Link had issues from the start (were reported mostly by red-ray several times and with each release he posts them anew). Some unfixed issues are dated 2014....

 

Honestly I have a feeling that red-ray fixed/helped more that the Link-Developers and each and every Corsair-employee that ever accessed this forum altogether...

 

I (and many others) won't touch Corsair products with a six meter pole thanks to this bad (software) support.

 

Fair enough...

Admittedly, I have neither the time nor the patience to review posts spanning 3+ years (I have reviewed the numerous summary posts). And I understand that was not likely the meaning of your reply but, rather, in my own interpretation, there have been some outstanding software defects that need to be resolved. I will take my own angle of pursuing the Corsair monoliths that stymie development, quality control, and trust within an organization. I don't care if I get banned. Someone needs to be held accountable for failed product launches and it is not engineering and, likely, not, product management. This must be systemic. There must be an executive that bases nothing in reality and forces "poo" on others.

As a software engineer myself, I do not want to release something I know is not ready. I tell that to my manager and director and they stand behind me. It's not just me, either. Same goes with my co-workers and QA! If we're not ready, we don't release. One or more customers may be pissed at the delay; they are much happier with a reasonable delay than an all out failure.

Corsair does not seem to know this.

I sent an email to their PR address, guess what? Nothing. I said my audience was one.

Link to comment
Share on other sites
red-ray

red-ray

Posted
Posted
As a software engineer myself, I do not want to release something I know is not ready. I tell that to my manager and director and they stand behind me. It's not just me, either.

 

The software engineers are not really up to the job. For example they used a semaphore when they should have used a mutex which is a basic schoolboy error. Eventually we got them to change from using the Global\CorsairLinkReadWriteGuard semaphore and to use a mutex which that called Global\CorsairLinkReadWriteGuardMutex.

 

They also messed up the locking semantics they specified and further the CL 4.6.0.86 locking is still not 100%.

 

The development is done by a 3rd party not by Corsair engineers. Obviously I have not seen the design documents and/or code, but from all the issues I can tell it's none too good.

 

I would post the e-mail address and linkedin profile of the manager, but they get filtered out, using Google it's trivial to find them.

 

The interlocking semantics are still incorrect and given I reported this on 09-Aug-2016 why? I also e-mailed the product manager again who did not even reply to all my e-mails. I suspect it's a one line change and can be tested using SIV 5.12 or later. It might have been half fixed, but when I open Configure Fan then pressed [Apply] a few times [Link Test] reported @ 00:29:50.894 seq 40 5 = CreateFile( \\?\usb#vid_1b1c&pid_0c02#7289_1.0#{3c5e1462-5695-4e18-876b-f3f3d08aaf18} ) Access is denied. so I expect the CL4 locking semantics are still incorrect in some of the CL4 code. See http://forum.corsair.com/forums/showthread.php?p=892949 for the details.
Link to comment
Share on other sites
unilythe

unilythe

Posted
Posted

There are reasons te believe that soon all of Corsair's products will start using just one application or at least will be linked together somehow. Right now, Corsair's hardware and peripherals use seperate applications, but I've heard for example OC3D hint that some time soon you'll be able to link the lighting of all those products together. I want to emphasise that he was hinting and I may have misinterpreted it.

 

It's just speculation in any case, but I'd love it if Corsair Link would just be thrown in the trash, and CUE would become the one application for everything.

 

The video with timestamp where this was mentioned:

 

[ame]

[/ame]
Link to comment
Share on other sites
red-ray

red-ray

Posted
Posted
The video with timestamp where this was mentioned:

 

So what will be done? I suspect the CUE GUI will talk to the same old CLink4Service and none of the issues within the service will be addressed.

 

This all sounds rather familiar...
  1. CL2 -> CL3 will fix all the issues...
  2. CL3 -> CL4 will fix all the issues...
  3. CL4 -> CUE will fix all the issues... :thinking:

Link to comment
Share on other sites
unilythe

unilythe

Posted
Posted
So what will be done? I suspect the CUE GUI will talk to the same old CLink4Service and none of the issues within the service will be addressed.

 

This all sounds rather familiar...

  1. CL2 -> CL3 will fix all the issues...
  2. CL3 -> CL4 will fix all the issues...
  3. CL4 -> CUE will fix all the issues... :thinking:

 

Yeah, I'm not saying it'll fix all the problems. I just hope it will. It's very possible it will only "link" the two together which would probably even make the issues worse rather than fixing them. I'm just cautiously optimistic that Corsair knows of the problems and genuinly aims to fix them, but that's probably a little naive of me. The actual chances of Corsair Link being dropped entirely and all functionality being added to CUE instead is pretty low, I guess.

 

Then again, the only issue I've really encountered in my small week of using Corsair Link is the ridiculously high CPU usage. I'm sure only a week is by far not enough to notice the bigger issues though.

Link to comment
Share on other sites
red-ray

red-ray

Posted
Posted
Corsair Link is the ridiculously high CPU usage.

 

To address this I expect Corsair would need to rewrite the CLink4Service in C/C++ and redesign the communication between the service and the GUI.

 

Using C## for the service in the first place was a poor decision as was using a vast number of TCP packets for communication.

Link to comment
Share on other sites
unilythe

unilythe

Posted
Posted
To address this I expect Corsair would need to rewrite the CLink4Service in C/C++ and redesign the communication between the service and the GUI.

 

Using C## for the service in the first place was a poor decision as was using a vast number of TCP packets for communication.

 

I actually program in C# as my job, and I can assure you that if done properly, it's perfectly possible to write a good C# application that has regular CPU usage. In most situations nowadays, C# is only about 10% less efficient than C++. In some very specific and limited cases, C# is actually more efficient because of JIT optimizations. C++ is generally more efficient, definitely, but the difference is minimal nowadays.

 

What I mean to say: C# isn't an excuse, honestly.

 

TCP packets I don't have a lot of experience with, but I can see how that could easily cause high CPU usage if not written properly.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing
×
×
  • Create New...