I've heard this mentioned a couple of times, especially in relation to the likes of F8W T5 bulkhead fixtures.  Specifically that it seems that newer halophosphate lamps have lower outputs than the older ones did.  Today, I believe I have encountered evidence of this first hand, when firing an old, albeit un-branded F8T5 Daylight tube up.  This sucker is BRIGHT in comparison to any other 8W tubes I have.  Unfortunately I don't have any modern daylight tubes to compare to, and realise that they are generally somewhat brighter for a given wattage than the other colours anyway - but the difference between this and the likes of a GE F8W/33 is quite staggering - The 33 is bright, but you can reasonably comfortably look at it from a couple of feet - the older daylight lamp is downright uncomfortably bright at that distance.

I was quite astonished to compare the current lumen figures for a Daylight F8W lamp, which seem to be around the 350Lm mark depending whose datasheets you look at, to a rating of 400 (GEC 1975 Catalogue, and a couple of others which I can't remember now).

Is it true that we've really lost a full 50 lumens from these lamps?  If so why?  I'm assuming changes to the phosphor composition and guessing for environmental reasons.  Would be interesting to see if that could be confirmed though.

I can tell about the 36W T8 from 3 manufacturers, from mid 2000s (assume 2005)

Philips (good quality)
Eurolux (medium quality)
Luxten (bad quality)

Lamps rated :
Philips - 2500Lm
Eurolux - 2700Lm
Luxten - 2500Lm

Initially Philips and Eurolux are bright, Luxten maybe a bit less

After some use, Philips and Eurolux ok, Luxten dropping in brightness

After some more use, Philips remain ok, Eurolux phosphor seems ok but it appears to begin to go mercury starved, Luxten either even darker or EOL

After some more use, Philips remain ok, Eurolux phosphor seems ok but its a bit mercury starved, Luxten EOL

Thanks for the response Ash, what I'm really referring here is that there seems to have been a significant drop off in the NEW rated lumen output of these tubes, rather than the rate of degradation.

The F8W one I've got here has got at least some hours on it, and is *far* brighter than any of the other new lamps I have - I really need to pick up a daylight one though to do a side-by-side comparison more accurately though.

According to the datasheets, a difference of 50Lm "out of the box" is an appreciable difference - 12.5% if my maths is correct.

I dont think the ratings are accurate to begin with, and can vary between manufacturers. In my example, the Eurolux (which can be expected to be in the same quality class as the older T5 you have) has a bit higher rated output, though i did not see it side by side with the Philips and its hard to tell whether it really is brighter

The Philips T8 seems to be 2500Lm from the beginning (the 1st ones i seen from the mid 90s). I can't compare it to older T8's (didnt see any), but the 40W T12s were rated 2300-2500Lm

Do not forget, then the rating mean lumen output after 100hours of burn-in. And with more and more tube types it become stressed in the datasheets, then the 100hour burn-in is essential to achieve the rated lumen output. I would guess it is related to the mercury dosing techniques using amalgams, where the burn-in hours may be really necessary for complete mercury release.

Are the new lamps made in China? I'm wondering if phosphor quality is getting spotty these days. I've come across a few that were terrible.

New Osrams and Sylvanias are China

We have in the bathroom of my mother home, a 2X8W T5 HF fixture. This fixture have now 2X8W LUXTEN T5s.
Each time i turns the light on, they come on like this: Picture
After a minute, returns to normal 54/765 daylight halophosphors color.

I guess it is the lack of evaporated mercury, so only the argon generate the UV for the phosphor. And as the Argon have no green/yellow lines as the Hg does (so it is missing in the spectrum), the lamp color is different (beside the low brightness).
After the lamp warm up, the Hg evaporate.

On electronic this take longer, as the real power delivered to the lamp is lower then on the magnetic (the lamp have higher efficacy on HF) for the same (so e.g. nominal) light output (so ballast factor). And many electronic ballasts use even lower power for the lamp (unlike the magnetic preheaters, on electronic the lower current does not complicate the lamp ignition).

Medved: This problem is unique to these LUXTENs, and there are still mercury, as the lamps are still bright in cold starting and as the mercury is in the spectrum.
If only the argon was energised, these lamps would simply glows a dim purple color (Mercury starved).

Guess : the lamps might be too short to have the "classic" mercury lack pattern (brighter in the ends) as the temperature gradient on the lamp switched off (so - mercury condensation pattern) is not so harsh, or a similar effect

Medved: This problem is unique to these LUXTENs, and there are still mercury, as the lamps are still bright in cold starting and as the mercury is in the spectrum.
If only the argon was energised, these lamps would simply glows a dim purple color (Mercury starved).

Isn't the purple color linked to the tri-phosphor?
Because the argon does emit UV in similar range as Hg, so phosphors respond to it, so radiate by it's colors. The Ar emit similar blue as the Hg.
So compare to normal Hg discharge, there is lack of only the green and yellow Hg lines, rest would be about the same.
Triphosphors are separated narrow lines, controlled rather separatelly, so designed to better complement the Hg radiation to form the nice, high CRI white. And with Argon only, the green and yellow Hg lines are missing, disturbing the balance, so the purple tint as the result.

As Halophosphates are quite continuous spectrum (peaking at green, sloping down towards red and blue), I doubt it would create an effect of purple color - I would expect on Ar discharge only a bit warmer tone then normally.

Medved: There are no triphopsohrs in these LUXTEN 8W T5s, only halophosphors.
When i turns the lamps on, they emits this color, after a minutes returns to a 54/765 daylight halophosphors color.
If only the argon would be energised, they would be produced the color of it.
The lamps are still bright enough at cold start, and the mercury lines are still pesents.
Also, halophosphors have peaks in the blue and the yellow, not in the green, and are lack in red, hence their poor CRI.

I didn§t mean sharp peak, but the shape of a "hill" spread over large part of the visible spectrum and centered around the green. The radiation in red is indeed very weak, it is only "tail" of that bump.
Are you sure the halophosphate phosphor have sharp peaks? I have more an impression, then the sharp peaks in the halophosphate lamp spectrum are in fact "original" mercury lines and not the phosphor...

And the CRI~70 could be even result of the tri-phosphor mix, if it is optimized for efficacy (so more green, yellow, on account of the red).
Unlike with halophosphates, with tri-phosphors you have nearly full freedom to choose the relative heights of different peaks, as in most cases each phosphor component create only one strong peak in the spectrum. In halophosphates the single component create the nearly whole wide spectrum "hill"
Normally the phosphor component's ratio in lamps is tuned for the best CRI (yield 80..90 for 2500..5000K, it goes down for higher color temperatures), but you may tune them to sacrifice the CRI in order to improve the efficacy (radiating the given power around green yield more lumens then radiating the same in red, but in the first case the weaker red could cause lower CRI)

The Luxten 8W T5 lamps are daylight halophosphors. They have the same specturm of the Philips TL-C 40W/54 in my brother room at my mother home.
The peaks are of the mercury lines.