E-bin P7

[uk_caver]

P4-U for example started out 91-118.5, which consumers obviously thought they had an equal chance of buying form any portion of that bin.

Obviously?
Even without baterija's contribution, I'd have thought that it was fairly obvious that if there are no devices available in bin N+1, then it was deeply predictable that in bin N, the outputs would be biased towards the bottom end, since even without knowing any production details, there's highly likely going to be some kind of bell-curve-ish tail-off happening.

To be honest, I'm not particularly bothered if people who can't even see *that* feel misled, since eternal thoughtless optimists are likely to make a habit of being misled almost anytime they buy anything.

For anyone not obsessed by numbers, surely the most important thing about a bin is whether there's any obvious difference between devices at the low end and high end of a given bin. Even with the old Seoul bins, the best and worst in a given bin are still pretty indistinguishable for practical purposes.

When it comes to Seoul splitting some bins, maybe they were just learning a marketing lesson from Cree - after all, there are some people who'd turn up their nose at a Q5 once R2s were out, even though the difference in practice is meaningless, though I suppose that's partly down to people wanting to be able to talk about having an R2-based light, rather than actually using it.

...the best conclusion we can draw is that SSC bins are purposely layed out so that the largest possible % of production falls in the bottom of that bin.

Since, apart from the splitting of S, T, and U bins, the current Seoul bins up to bin X seem to be virtual copies of the Luxeon bins from years back, right down to the letters used, I guess you must credit the Lumileds engineers with a scary amount of foresight, predicting where to draw the T-U boundary for best effect for a future competitor.

Personally, I greatly prefer being able to buy LEDs of known Vf in bins that give an adequate idea of brightness than have LEDs in tighter brightness bins, but where Vf is a total gamble.
Other people may want something different, and they have a company that caters for them.
To me, that seems like a good thing.

 

[baterija]

Obviously?
Even without baterija's contribution, I'd have thought that it was fairly obvious that if there are no devices available in bin N+1, then it was deeply predictable that in bin N, the outputs would be biased towards the bottom end, since even without knowing any production details, there's highly likely going to be some kind of bell-curve-ish tail-off happening.

Sure way to be more clear than me and show me up. :twothumbs

If anything Seoul's practice of splitting the bin gives more clues to shifts in the distribution curve than fixed bins. Cree R2 are starting to be seen with more frequency lately than they were for months after they were first released. That's a clue the curve is shifting right slowly but less direct.

Cree's binning is also almost linear. That means diminishing returns in perceived brightness with each step. I could label that as dishonest marketing. It is what it is. Seoul bins more according to perceivable difference, and then chooses to split the bins. That is what it is. We're all smart enough, and have enough info here to make informed buying decisions with both binning strategies. If we choose not to, shame on us.

I'm still waiting for Natural White 93 CRI P7's, even if they could only manage A or B flux bins. :naughty:

 

[saabluster]

If anything Seoul's practice of splitting the bin gives more clues to shifts in the distribution curve than fixed bins.

Nobody here is questioning the shift in the distribution curve. Of course they are getting(not making) brighter LEDs because Cree's dies continue to improve.The big question that was so conveniently overlooked was "why would SSC be about the business of bin shifting?".

Cree's binning is also almost linear. That means diminishing returns in perceived brightness with each step. I could label that as dishonest marketing.

Seriously?:shakehead

Seoul bins more according to perceivable difference,

Since people seem to want hard evidence and certified lab tests in this thread I am afraid I am going to have to ask you to back that up.

and then chooses to split the bins.

So let me get this straight. They, knowing full well that the lumen march will continue on, binned according to perceivable difference and then for some unexplained reason chop bins into smaller pieces? Doesn't this rail against their entire way of binning? If as you say they bin according to percieved differences they would be increasing the size of the bin as the distribution curve moves to the right not decreasing.
Heres the thing. They know full well that outputs will continue to increase. They can figure out ahead of time what a good binning system is and stick with it. Why fiddle with the bins?

Evidence as to their thinking and methods
It is well documented that they are disingenuous in their press releases.
Take a look at a clip from one of them from here.
12-04-06
Seoul Semiconductor's P4 LED

SEOUL, Kr. - Seoul Semiconductor (SSC, KOSDAQ 046890), a global leading LED manufacturer, announced today the launch of its new LED product, P4. The P4 emits 240 lm at 1 A of light and features the industry’s highest luminous efficacy (100 lm/W @ 350 mA at the maximum) with only a single die, making it a brighter and more cost-effective light source compared to conventional 70 lm/W fluorescent and 15 lm/W incandescent lighting options.
“P4 is making history in the lighting industry as the alternative to conventional fluorescent and incandescent lighting sources,” said Jung Hoon Lee, CEO of Seoul Semiconductor. “Conventional LEDs have been known to emit more than 100 lumens with several dies. Seoul Semiconductor’s P4 is the only LED product in the world with 240 lm via a single die.”

Notice how they word it to sound like if you ordered a P4 you would be getting an LED that put out 240lm at 1A?:crackup: Maybe one out of a hundred in that bin would.

About six months later Cree put out a press release for their 100lm/W LED. Notice the difference in tenor.

"Cree...today announced commercial availability of XLamp® LEDs with minimum luminous flux of 100 lumens at 350 mA. XLamp LEDs are the first LEDs to be available in volume with this level of performance."

Whereas SSC boasts about the output of the very top of their top bin Cree showcases the minimum expected.
Notice at the end there they acknowledge the prior "availability" of LEDs of this caliber but note that they were not available in an significant number.

Since I have been watching this scene Cree has always been above board with their press releases not trying to word things to make themselves sound better than they are. The same cannot be said about SSC. Yes I have only sited one reference for them but you are welcome to search through all the press releases and find all the other instances. Of course you will have to negotiate around all the news about them and Nichia suing one another.

If SSC has a track record of being deceiving with the way they deal with the public this is evidence as to the corporate culture there. Who decides binning? Who sits down and figures out the business strategy? Who figures out how to squeeze more money out of an operation?
Why be shifty with the bins? Why have the SSC LEDs tested around here by highly respected members barely made it in their bins. This is simple math for all who want to be honest with the available evidence at hand. Would more evidence be better? Absolutely! But so far the evidence leans against SSC and until any evidence can be shown to the contrary I cry foul.

We're all smart enough, and have enough info here to make informed buying decisions with both binning strategies. If we choose not to, shame on us.

No "We"(humans) are not all smart enough. This is in evidence every day even here on an enthusiast website. We(CPFaholics) are freaks. Normal people do not know this much about LEDs and that includes a lot of companies that would be in the market to use LEDs in their products. And even if nobody was ever duped by the misleading practices of SSC the fact that they try ticks me off and should give some clue as to the people running it and the possible reasons why they would be shifting bins at such "convienient" times.

 

[uk_caver]

No "We"(humans) are not all smart enough. This is in evidence every day even here on an enthusiast website. We(CPFaholics) are freaks. Normal people do not know this much about LEDs and that includes a lot of companies that would be in the market to use LEDs in their products. And even if nobody was ever duped by the misleading practices of SSC the fact that they try ticks me off and should give some clue as to the people running it and the possible reasons why they would be shifting bins at such "convienient" times.

The average retail customer likely still buys lights because they have "3W LED!!!" printed on the packaging in bright colours, even if the LED isn't likely to see anything like 3W in their life, and might even still be a Lux III, and even if the customer has no idea what a Watt is.

If it comes to a company buying LEDs to use in products, if they're too dim to read a datasheet or make a vaguely educated guess about binning, and they confuse maximum, typical and minimum, or assume that there's a linear spread of devices across each bin, then they shouldn't blame anyone else for that.

However, in practice, it doesn't matter what the spread is across a bin, as long as a bin is narrow enough that the variation from top to bottom isn't large enough to make a difference for the given application.

Personally, with both the (unsplit) Seoul bins, and the similar-width Luxeon bins they were copied from, I never had any worries that if I made a load of lights from devices in the same bin that anyone would notice (let alone worry or complain about) any brightness difference between their lamp and anyone else's.

 

[Ryanrpm]

There's no way I'm able to contribute anything to this thread, but just wanted to say, "Nice comparison of press releases saabluster." It's nice knowing what is really out there, and not just what someone thinks they know is out there.

 

[baterija]

Nobody here is questioning the shift in the distribution curve. Of course they are getting(not making) brighter LEDs because Cree's dies continue to improve.The big question that was so conveniently overlooked was "why would SSC be about the business of bin shifting?".

I'm simply stating that the bigger bin followed by a split gives me more information about where the curve lies in relation to the bins than Cree's method when changes are happening slowly like they currently are. It's certainly not what I would encourage as a best business practice if I had input at Seoul. I didn't overlook the question of why; I simply don't care very much. I'd care if Seoul was consistently grading LED's in a bin they don't perform in.

Cree's binning is also almost linear. That means diminishing returns in perceived brightness with each step. I could label that as dishonest marketing.

Seriously?:shakehead

No, spin is great isn't it. :naughty: That just may be why I used the conditional phrase "could" in there. I'm perfectly serious about the diminishing returns in their single die binning structure though. Anyone buying an R2 at a premium because it's "brighter" than a Q5 might agree with the dishonest notion when they can't tell the difference.

Since people seem to want hard evidence and certified lab tests in this thread I am afraid I am going to have to ask you to back that up.

Cree's binning structure doesn't need to be tested - they define it and publish it. Look at the percentage gap between the top and bottom of their single die bins from P2 - R2, it's diminishing.

So let me get this straight. They, knowing full well that the lumen march will continue on, binned according to perceivable difference and then for some unexplained reason chop bins into smaller pieces? Doesn't this rail against their entire way of binning?

It does. Maybe it's dishonest. Maybe they've got some organizational dysfunction that pulls them between 2 binning solutions. (Somewhere there could be a Seoul Engineer that looks like Dilbert with his very own pointy haired boss overriding reason. Maybe Wally is in charge of binning.) It simply doesn't matter to me. I got involved in this thread only because I wince at any notion that top bins are not weighted towards the bottom.

Or more simply, I don't have a dog in this fight.


Buy both!:sssh:

 

[saabluster]

Cree's binning structure doesn't need to be tested - they define it and publish it. Look at the percentage gap between the top and bottom of their single die bins from P2 - R2, it's diminishing.

I was not talking about backing up Cree binning. I was talking about SSC's purported binning to perceivable differences. Go back and read it again and you will see what I was talking about.

I wince at any notion that top bins are not weighted towards the bottom.

I agree whole heartedly with you there.

 

[uk_caver]

As to perceptible differences between LEDs of slightly different brightnesses, a lot depends on the application.

I can see that maybe for some products using LEDs, it may be that tight binning is needed (or at least specified/desired by the end user), but for flashlights/headtorches being used in practice, I doubt that many people would reliably notice a +25/-20% lumen difference in the absence of some kind of reference.

I'm sure it's possible to distinguish smaller differences in artificial comparison setups (lights mounted side-by-side, rapid alternating switching or covering/uncovering of lights), though I think you'd find that if there was a pause between light A going off and light B going on, it would make it harder to tell, and the longer the pause, the harder it would be.

If you gave someone one of two lights and got them to walk through a dark scene using it, and then repeat that walkthrough with a second light, and asked them what difference they thought there was, if any, I suspect that many wouldn't be sure that there was a difference until it was somewhere in the ballpark of a Luxeon/old Seoul binwidth.

 

[easilyled]

This thread seems to have become completely derailed from the OP's original question about when we can expect to see SSC-E-bins
and hijacked into personal gripes about Seoul instead.

My aim was to contest the original reply to the OP's question by Marduke, which was "probably never"

I didn't see how anyone could make that sort of prediction without recourse to some more factual basis such as measuring
the output of the existing highest bin (D-bin)

I wouldn't be surprised to find that most D-bins are nearer to 800 than 900 lumens, but I would rather confirm this than speculate.

At some point in time, there is likely to be an incremental increase in output due to some technological advance.

At this time the flux of the D-bin will likely go up and then an E-bin will probably be announced soon afterwards.

It seems to me that without measurement of flux, this point in time will be impossible to predict.

I think however that the fact there are now D-bins with a Vf of "I" means that some advance has already been made
compared to when only J Vfs were available if I had to resort to speculation.

Anyway, I couldn't care less whether a particular bin is weighted to the lower end or not.

I have many flashlights with Seoul leds and many with Crees and I'm very happy with the performance of both of them which
seem to me to be on a par with each other in different ways.

I really think that the people that have personal issues with Seoul should have resisted the urge to air their views in this thread
as it wasn't relevant to the OP's original question.

If you feel so strongly, why not make a new thread about this instead?

 

[saabluster]

This thread seems to have become completely derailed from the OP's original question about when we can expect to see SSC-E-bins
and hijacked into personal gripes about Seoul instead.

My aim was to contest the original reply to the OP's question by Marduke, which was "probably never"

I didn't see how anyone could make that sort of prediction without recourse to some more factual basis such as measuring
the output of the existing highest bin (D-bin)

I wouldn't be surprised to find that most D-bins are nearer to 800 than 900 lumens, but I would rather confirm this than speculate.

At some point in time, there is likely to be an incremental increase in output due to some technological advance.

At this time the flux of the D-bin will likely go up and then an E-bin will probably be announced soon afterwards.

It seems to me that without measurement of flux, this point in time will be impossible to predict.

I think however that the fact there are now D-bins with a Vf of "I" means that some advance has already been made
compared to when only J Vfs were available if I had to resort to speculation.

Anyway, I couldn't care less whether a particular bin is weighted to the lower end or not.

I have many flashlights with Seoul leds and many with Crees and I'm very happy with the performance of both of them which
seem to me to be on a par with each other in different ways.

I really think that the people that have personal issues with Seoul should have resisted the urge to air their views in this thread
as it wasn't relevant to the OP's original question.

If you feel so strongly, why not make a new thread about this instead?

Don't try and lay this at my feet. As my mom would say it takes two to argue.:poke: Its not like the topics in this thread are completely separated from from the OP's topic. "when is SSC releasing the E bin P7?" Because of SSC's practices we don't know. They may just shift bins tomorrow to move the E bin down to the level they can produce at. That has everything to do with the OP's question. Now a better question would be "When is SSC releasing P7s at at 900 or 1000 lumen output?". But that wasn't the question.

 

[Marduke]

No rebuttle to all this TexLite??

 

[easilyled]

Don't try and lay this at my feet. As my mom would say it takes two to argue.:poke: Its not like the topics in this thread are completely separated from from the OP's topic. "when is SSC releasing the E bin P7?" Because of SSC's practices we don't know. They may just shift bins tomorrow to move the E bin down to the level they can produce at. That has everything to do with the OP's question. Now a better question would be "When is SSC releasing P7s at at 900 or 1000 lumen output?". But that wasn't the question.

Your replies were obviously agenda driven. Mine weren't. :poke:

I understand that you don't like the way that Seoul have changed their bins lately. I think everbody's got the message now, so well done!

Personally I'm very happy with 800 lumens output.

I have several lights with SSC-P7s and the output is amazing.

 

[saabluster]

Your replies were obviously agenda driven. Mine weren't. :poke:

I understand that you don't like the way that Seoul have changed their bins lately. I think everbody's got the message now, so well done!

Personally I'm very happy with 800 lumens output.

I have several lights with SSC-P7s and the output is amazing.

My only "agenda" is finding the truth. I also did not like everyone hounding Marduke. And agenda or not you participated and perpetuated the discussion.
Yes I prefer Cree over SSC and for very good reason but I also own many SSC products. One of my favorites are SSC's high CRI line of LEDs that Cree as yet has no match for.

 

[uk_caver]

Its not like the topics in this thread are completely separated from from the OP's topic. "when is SSC releasing the E bin P7?" Because of SSC's practices we don't know. They may just shift bins tomorrow to move the E bin down to the level they can produce at.

The only time Seoul have shifted (rather than merely split) bins seems to have been very early on in the P7's life, before most people actually got their hands on one. It appears the P7 was announced as available at the very end of February 2008, and the bins had been changed to their current ones by some time in April.
If they'd done it much earlier, it would barely have counted as a change at all.

It would seem very much less likely that they (or anyone else) would change bins once production and consumption was in full swing.

 

[ECKO32]

Hi,
Just checking out the Bin coding charts for the P7 & notice that there are quite a few different P7's. Why is it that an LED can have so many different tints, VF's & Brightness groups?
I've only seen C & D bins available, when is SSC releasing the E bin P7?
Cheers
Justin

Well i guess this threads not about the e-bin anymore such a pity politics has to step in and take it over;(
sometimes its so distracting when I have to hyperlink to another thread but its worse now that the tread strays off topic wish you guys would start a new thread regarding this if it be cree vs. ssc or whatever:(:(:shakehead

 

[Nil Einne]

Hopefully people don't kill me for resurrecting this thread but it's always interesting to look at things in perspective. AFAIK, so far, still no E bin P7. But also no splitting the D bin either as some people predicted. Whether this is because they are still only performing in the bottom half of their bin as some people allege I can't comment on.

 

[bullettproof]

Hopefully people don't kill me for resurrecting this thread but it's always interesting to look at things in perspective. AFAIK, so far, still no E bin P7. But also no splitting the D bin either as some people predicted. Whether this is because they are still only performing in the bottom half of their bin as some people allege I can't comment on.

LOL this thread is hilarious :crackup:

How those E bins treating ya:thinking::crackup:

 

[znomit]

LOL this thread is hilarious :crackup:

How those E bins treating ya:thinking::crackup:

To be fair the XRE hasn't progressed at all either... :poke:
:popcorn:

 

[saabluster]

To be fair the XRE hasn't progressed at all either... :poke:
:popcorn:

As I have come to find out that is because Cree is intentionally holding the XR-E back. They are trying to move customers over to other more profitable product lines. I was told not to expect any improvements on the XR-E platform. Interesting isn't it that the XP-E which uses the same dice as the XR-E has an R3 bin in volume availability?

 

[znomit]

Interesting isn't it that the XP-E which uses the same dice as the XR-E has an R3 bin in volume availability?

I always assumed the ring sucked up a bin full of lumens.

I see cree still don't list R2 on XRE...