Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[EddieZ]

You seem to flaunt with your certifications and experience. Well, that does not make you a 'good' IT guy...
But anyway, comparing with a Cisco PIX is like comparing a Ford sedan (consumer) with a Mack (PRO).

Your claim as to the stabilty of 1.21...I do not know where you did your testing, but it is regarded by the majority as the worst firmware after 1.11... So hearing your qualification of 1.21 is surprising to say the least.

The issues reported are incidents. As an experienced IT guy (just following your logic about certifications for a moment) you would know that 5 complaints do not make a priority issue. Certainly not when you have no insight in the local configs. A malfunctioning restoration of saved config is quite common (see BIOS update manuals) and it is a preferred method to config from scratch (factory default). Solving issues is not about the odds. Refusing to look in different directions because of the odds does not suit an IT pro. Because the best thing in IT is simplicity: Either the firmware causes issues which are generic (all equal devices) and should be reproduceable or there are other factors causing the issue, not being the firmware.

Example: NTP server issue. I cannot reproduce it at home. The given NTP server just works fine. No reboots, no loops. Howcome? If it's the firmware I should be able to reproduce.

Example 2:
The only issue I could reproduce was reboot loops after restoring a saved config.

So thanks for your elaboration on...well, mainly yourself.

Let's try to fix this guys issue.

Generic steps would be:

1. reflash 1.3x
2. Reconfigure manually after setting tho factory defaults.
3. reset your internet gateway device

[sandman]

Your claim as to the stabilty of 1.21...I do not know where you did your testing, but it is regarded by the majority as the worst firmware after 1.11... So hearing your qualification of 1.21 is surprising to say the least.

The issues reported are incidents. As an experienced IT guy (just following your logic about certifications for a moment) you would know that 5 complaints do not make a priority issue.

We upgraded to 1.21 and we run a macbook pro, a PC with XP SP3, Canon MP wireless printer and Ipod Touch all problem-free since the upgrade. Before we had 3-4 dropouts a day. My only need to upgrade is for a version that supports shareport and macbook pro. See the Mac thread for problems there.
We recently swapped companies from a 6 Mbs ADSL service to a 20 Mbs Cable , so this firmware has been running on different services with different modems, speeds etc.
Lastly, if I check on D-Link Germany, 1.21 is the most up to date official firmware. Not 1.22b, not 1.30 or 1.31. The official seal of approval by D-Link Germany is for 1.21, so if it is so bad as you say, why has it not been returned to beta, as 1.30 was so quickly, If the "majority" have so many problems with it?

[TommyLee]

You seem to flaunt with your certifications and experience. Well, that does not make you a 'good' IT guy...
But anyway, comparing with a Cisco PIX is like comparing a Ford sedan (consumer) with a Mack (PRO).

Your claim as to the stabilty of 1.21...I do not know where you did your testing, but it is regarded by the majority as the worst firmware after 1.11... So hearing your qualification of 1.21 is surprising to say the least.

The issues reported are incidents. As an experienced IT guy (just following your logic about certifications for a moment) you would know that 5 complaints do not make a priority issue. Certainly not when you have no insight in the local configs. A malfunctioning restoration of saved config is quite common (see BIOS update manuals) and it is a preferred method to config from scratch (factory default). Solving issues is not about the odds. Refusing to look in different directions because of the odds does not suit an IT pro. Because the best thing in IT is simplicity: Either the firmware causes issues which are generic (all equal devices) and should be reproduceable or there are other factors causing the issue, not being the firmware.

Example: NTP server issue. I cannot reproduce it at home. The given NTP server just works fine. No reboots, no loops. Howcome? If it's the firmware I should be able to reproduce.

Example 2:
The only issue I could reproduce was reboot loops after restoring a saved config.

So thanks for your elaboration on...well, mainly yourself.

Let's try to fix this guys issue.

Generic steps would be:

1. reflash 1.3x
2. Reconfigure manually after setting tho factory defaults.
3. reset your internet gateway device

EddieZ, ...where to begin,

No one is flaunting here. D-Link needs to know there are users out here that are actually IT professionals in the industry. Not just hobbyists at home tweaking things in there spare time, whose day job may not even be IT. And in knowing so, realize that we (the IT professionals), will not be lulled over by simple statements like thousands of 655's have been sold, and there are only 10-problems out there (supposedly), so it must be the user and not the firmware. Even if the only problem that existed was the reboot issue you spoke of, and it could be completely cleared by starting from scratch as you say, a problem still exists that should be rectified.

Additionally and based on your response, it would seem that everyone else in the forum is inept at successfully configuring a working DIR-655, all because your router works with no problems according to you. Where is the logic in that? Better yet, where is your so called IT professionalism in that?

The bottom line is, v1.31 does have problems. Be thankful you got the luck of the draw with a combination hardware revision/firmware installation that luckily enough didn't happen to cause you any issues. However again, just because your router works according to you, it in no way means that firmware v1.31 is utterly and totally without issue. I have never worked for D-Link personally, but I have worked in QA throughout my career, and I am well aware of how things are pushed through to make deadlines. If known dev issues are considered to not be fatal, or show stoppers so to speak, then the company pushes forward with the changes hoping for the best, or at the very least, to deal with them another day. The only problem is far too often, those non-fatal issues are never revisited and or resolved, leaving the customer to simply deal with them.

I also agree that we should try to help people in this forum, like "desides" for example. However advising them to perform the same repetitive steps over and over, I guess until they eventually work? ...does not constitute helping anyone in my opinion. Especially when the only true help is going to be a rectified revision of the firmware, or at least an unlock revision to allow down grading back to v1.21.

Respectfully,

Tommy Lee

[ledan]

That's cutting corners and a very simplistic ('easy') reasoning.
Question: did you restore your 1.2x config in the new 1.3x or did you reset to factory default and rebuild the configuration?

I second the opinion that three complaints do not make a faulty firmware. My 1.3x works great but in your line of reasoning I should have the problem too?
Again people seem to think that 1.11 is the holy grail of firmwares. Why? Because they think that the measurements/gauges in 1.11 were more accurate than the those in the newer firmware, thus showing better connection rates. It's the other way around: the 1.11 is more inaccurate (too optimistic). There has been no change in radio strength. The question has been put over and over again at Dlink tech: they made no firmware changes concerning radio strength...

I used measurements on my laptop in the same room and there was a real difference in downloading speed while navigating between 1.11 and 1.22 on A2 hardware.

[EddieZ]

I used measurements on my laptop in the same room and there was a real difference in downloading speed while navigating between 1.11 and 1.22 on A2 hardware.

I measured and tested too. 1.22 gave exactly the same throughput as 1.11, but had to reconfigure the antenna setting somewhat. The radio seems a bit more sensitive after 1.11 which changed the optimal position. 1.31 is the best so far with actual throughtput and transfer speed.

[ledan]

I measured and tested too. 1.22 gave exactly the same throughput as 1.11, but had to reconfigure the antenna setting somewhat. The radio seems a bit more sensitive after 1.11 which changed the optimal position. 1.31 is the best so far with actual throughtput and transfer speed.

Thanks for the information. May be I'll try 1.22 again to test antenna position before moving to 1.31

[Nnyan]

Could anyone that was having issues with 1.22 dropping during bittorrent downloads post if 1.31 fixed the issue?

[DaveH]

Here is my take; an average PC user. I had issues when I initially upgraded to 1.31NA. Thanks to this forum I learned the correct way to install and make it work, so I thought. No longer do I have consistent connection drops on either wired or wireless devices. However, my VoIP that worked flawless prior to upgrading to 1.31NA does not work. It cuts in and out and is now unusable.

As a test, I wired my VoIP directly from my cable modem (not even involving the DIR655 in the equation) and it worked perfect. I made a 30 plus minute call. Absolutely no quality issues on my end or the receivers.

I have been reading the comments form IT professionals and D-Link moderators going back and forth. It has been entertaining to say the least. Bottom line is since I upgraded to 1.31NA, my set up (cable modem to DIR655 then one hard wired PC and one hard wired VoIP) that worked prior is no longer working correctly. I have always done firmware upgrades. How silly of me upgrading to a firmware that was issued by the manufacturer to presumably improve and enhance their product. All while assuming it was issued so an average, non-IT pro could install and enjoy the benefits without spending hours on a forum to get it to work!

Ok, so I am having problems. I can understand stuff happens. I am sure D-Link will address these issues and make corrections eventually. But why the hell do they no allow the user the option to fall back to a prior proven firmware that worked in their specific case?!! They have eliminated an escape route. This is just plain ignorance. That is what frustrates me most. Yes they had a disclaimer written in red you cannot go back. But again, I trusted it would work. Lesson learned. I will now let others try it and what for feedback. 

I am not here to get into a pissing match. I just want my internet and VoIP to work properly! Does anyone have suggestions on how to make this happen? Can I go cable modem to VoIP then to DIR655? If so, are there any drawbacks. Guess I could go try that. Or maybe I'll go purchase a new non D-Link product. That would be a shame up considering I loved the way it worked prior to this mess.

[EddieZ]

Can you provide an overview on your QoS (and WiSh) settings?

[DaveH]

Whatever the default setting are, those are my settings. I have not played around with those settings in the six months I have had the router. If you still need specifics, I can will get that for you.

[teknomedic]

I didn't read every detail of all the posts, but for those having issues... have you attempted a re-download of the firmware and then another "30/30/30, flash, 30/30/30, manual setup"??  I'd suggest at least doing that again if you're having issues to maybe remove any corrupted data in the download or a possible "bad flash".

Also, as a general comment... not all hardware is equal and while I would think most people are doing just fine with 1.31 I would also think that some people would have issues with 1.31 as well.  Since I didn't write the firmware I can't say for sure, but maybe this firmware is stressing certain parts of the hardware differently over 1.21, 1.22b05, etc.

Computer hardware is an interesting thing.  While the same people who had issues with 1.21 are now singing the praises of 1.31... others that had no issues with 1.21 are now denouncing 1.31 for giving them huge headaches.

So (for the ones having problems and if you haven't already tried it) if it was me I'd try to re-flash my router again with 1.31... I'd also try 1.30 just as a test to see if the same issues were there.



**On a side note (I know we've been told the "reasons", but)... I'd like to know why exactly we can’t downgrade to an earlier firmware.  I of course don’t know what all the properties of the chips inside the router are, but if we’re talking any type of flash ROM that was obviously able to be changed to “lock out” a downgrade… then why can’t that same chip be changed again to allow a downgrade??  Is this a “write once” type of memory that was changed with 1.30/1.31 and now we’re locked to that code no matter what???

Please explain in detail exactly why we can’t downgrade… because from everything I know about hardware there should be no limitations so it VERY much appears to be an artificial software limit imposed by D-Link even though the hardware could still be “re-flashed” with the old code.

[EddieZ]

Whatever the default setting are, those are my settings. I have not played around with those settings in the six months I have had the router. If you still need specifics, I can will get that for you.

Default is "no settings".

And for teknomedic: the 1.3x firmware has a new SDK or Kernel update. Once installed, it locks the new codebase (kernel) in the device. Since the older firmware requires a different kernel (the old kernel) it is not downgradeable. Can't go into more details.

Ofcourse one can say "there is always a way..." but believe me, you don't want to go that road.

[teknomedic]

And for teknomedic: the 1.3x firmware has a new SDK or Kernel update. Once installed, it locks the new codebase (kernel) in the device. Since the older firmware requires a different kernel (the old kernel) it is not downgradeable.

I understand that part... what I'm trying to figure out is just why the hardware can't be reflashed with the "old kernel".  Is it because it's a write once type of memory (which I wouldn't think as it would mean only one update like this could ever happen)??... or is it because the older firmware files don't include the full kernel update code (in which case a v1.21 with full kernel update could be released).

Simply saying that it's a new kernel doesn't fly with me (and others from what I've read) unless there's some physical reason the hardware can't be changed.

Let me put it this way... even though I'm running Vista and can't run anything from Win98... I could still reformat my hard drive and put Win98 onto my computer (a vastly older kernel).  So please tell me why we still can't reformat our "hard drives/routers" and put the v1.21 kernel back on?

[teknomedic]

I guess what I'm trying to get at is this:

IF D-Link simply wanted to lock us out from using anything older once we upgraded then fine... but then it's ONLY a software lock as I would assume all the chips in the router would be of the "rewrite or CD/DVD RW" type and not the "write once or CD/DVD R" type.  In this case I would much prefer D-Link to simply say... "because we didn't want you to use anything older so we locked you out".

BUT...

IF it's an actual physical limitation of the hardware (or earlier firmwares), then I'd like to hear the WHY and WHAT of the physical limitations.  (ie, write once flash, v1.21 or earlier don't contain the kernel update codes, etc).

....I've seen them do similar things with motherboard BIOS chips, where it won't let you downgrade, but it's just a software lock imposed and not an actual physical one as (I've done it in the past) I can simply remove the chip from my motherboard.... place it into my EEPROM programmer and re-flash an older BIOS and it works just fine.  So obviously there's not a physical limit, but a software limit imposed by the maker.

[ominator]

I guess with this new version they performed a full flash instead of a code flash, and flashed the bootstrap with a new version.

More then likely the old firmware flashers don't recognize the new bootstrap and refuse to flash it, or the old code does not work with the new strap.