Do you get an 'XP rating' applied when you o/c?

:cool: Basically do you find it somewhat dissatisfying if not downright annoying that when you o/c an AthlonXP and lose your XP rating (model number)?

:p eg. You o/c your TbredB XP1700+ from 11x133 to use 12.5x166 which is identical to an XP2600+ but then you get the deflating 'Athlon running at 2.08ghz' instead of the appropriate 'AthlonXP2600+' tag. I understand that you may run at a speed which doesn't have an XP rating such as 18x133 or 10x225 but surely it would be far nicer if you do run at a std XP rated speed for that speed to be displayed.

:( My old Gigabyte 7VAX KT400 mobo showed the XP rating when o/c'ed ... my Abit NF7-S v2.0 nForce2 just shows the '2.08ghz' ... okay it's not a big thing by any means but come on ... don't rob me of my rating LOL!

:D So when you o/c does your mobo apply the XP rating (if one is applicable) or simply display the ghz alone? It would be interesting ... but maybe I'm just insanely boring LOL!

Well, it goes without saying, but my Abit NF7-S v2.0 nForce2 also give a '2.2Ghz', instead of a 3200+

I'll try the AN7 when it arrives, and see. Doubt it will though...

2.31Ghz (tested prime for a day)

I know it is faster than even a XP3200+ so I don't mind "losing" the XP2500+ tag.

Okay, people who aren't too knowledgable might actually think that the 2500 means 2.5ghz and that I "downgraded". But as long as I know better ;)

Sisoft sandra gives you a guestimate if you run the CPU check....

Thats what id suggest using.

TiG

How does the XP rating work anyways?

Or does it mean to take, The Pentium 4 equivalent in mhz? Eg the XP 3200 is equal to a P4 3.2GHz. Thats the way I have always understood it. I'm up for correction ...

My XP1700's only at XP2500+, ie 11x166.

Need more spare time for fiddling :(

Austin mate..I think its a case of the BIOS recognising the chip type.

eg: Xp1700 overclocked to anything up to its max even with 166 fsb, shows as a proper overclocked chip type IF the BIOS knows that kind of chip exists....and to prove it....if you set the bus speed in the middle...say 150 ish..it only shows the clock speed.

If you take the muliplier DOWN below what was ever made, by AMD then it also doesnt have a "record " in the BIOS of AMD ever making that chip....

least thats how it works on my Asus A7N8X.....

one exception, ...cos PC's are desgined to break rules ;)....if I get a 1700Xp up to 12.5 x 166 its fine as an XP2600....but if the FSB goes even higher..it STILL shows as an XP2600....which is wierd...

Gonna shoot myself in the foot now....my NEW XP2500 (unlocked one, cos AMDare having my Locked one to "test" ) shows up in 3dmark 2003 as a DURON....:)


hehehe

NF7 used to do this, but it was changed due to popular request.
I want to know the exact mhz of the chip when i boot up, not some over inflated XP rating.
You can get the XP rating with the older BIOS's.
I also think i have another way of doing it, although im not sure yet, will have to check tonight.

Quote:

---

How does the XP rating work anyways?

Or does it mean to take, The Pentium 4 equivalent in mhz? Eg the XP 3200 is equal to a P4 3.2GHz. Thats the way I have always understood it. I'm up for correction ...

---

i thought it was the equivalent mhz of one of the old tbird athlons (ie a 1600+ was equivalent to a 1.6Ghz t-bird (if they'd made them - despide the xp being only at 1.43ghz).

it was introduced i think due to the fact that intel where whacking up the mhz of processors so customers thought "wow its 2.6ghz", but they didnt really increase the performance, hence an athlon of equivalent mhz speed would completely destroy a p4 in benchmarks (well apart from in sse2 tests)...

mark

I get a rating im apparently running a 3200+ at the moment :D

My Epox 8k9a2+ seems to think certain numbers can have ratings, and other ranges too, but there are a few blackholes where I just have speed and no rating. Doesn't really bother me, but it mucks up things like 3DMark a bit...

NS

Zak is more or less right - it's a case of the frequency determining the CPU Identification, or rather, contributing to it. There is a low-level function, CPUID, you ( or rather, software, including the BIOS) can call that returns a fair bit of identifying data, among which is the CPU ID String. This, in the cse of Athlons, takes the form :-

AMD Athlon (™) [xxxx] where xxxx is determined by a lookup table and the system frequency. The table reads (this is an extract but you'll get the idea)

Code:

---


MHz      Name

1333      1500+
1400      1600+
1467      1700+
1500      1800+ (Mobile chips at 200FSB only)
1533      1800+
1600      1900+
1667      2000+
1733      2100+
1800      2200+
2000      2400+
2083      2800+ (Desktop processors at 333FSB only)

---

AMD have some pretty strict controls over what must be displayed during boot up. For instance, with some processor models, if the BIOS just displays the frequency without at least ALSO displaying the proper ID string, including Model number, the motherboard will not get AMD validation and will not be included in their 'recommended' website.

So, it can be seen from this that if you have a weird combination of multiplier and FSB, the BIOS will NOT identify as the appropriate model.

Quote:

---

Originally posted by Austin
..... You o/c your TbredB XP1700+ from 11x133 to use 12.5x166 which is identical to an XP2600+ but then you get the deflating 'Athlon running at 2.08ghz' instead of the appropriate 'AthlonXP2600+' tag. .....

---

The AMD XP 2600+ can be either the Model 8 266FSB or Model 10 333 FSB. The 266 model uses 16 x 133, while the 333 FSB uses 12.5 x 166. It may well be that the model number is reporting one model and that does not match the FSB and multiplier combination for a 2600+.

More than that, though, the early model 6 Athlons (like the XP1700+) reported a CPUID of 662 but the model 8 and model 10's (including the 12.5x166 version of the 2600+) do NOT, therefore the BIOS CPUID function is presumably aware that even if it gets the right frequency, it KNOWS it is not a 2600+ which, at 166, would be a Model 10 and therefore not reporting CPUID of 662.

I hope that makes sense.

it does mate...nice post :)

if i underclock my CPU it will still read as a 1800+ for example, which was a Palamino origionally (if i remember correctly) so how can my barton be seen as a palamino?

Perhaps because the ID strings for OLDER processors are embedded in LATER ones, but it's hard to do the other way round :D

After all, Barton's were only a long-range plan when Palominos were released.

:) What I find annoying is the 'supposedly' best SktA mobo, the NF7-S v2.0 failing to apply the XP rating, it's no big thing but I find it strange that some mobos apply it (when it fits) while others do not. For example the Gigabyte KT400 I previously used gave an XP rating to any o/c'ed Tbred or Barton so long as the multiplier and FSB (hence CPU speed) matched. The NF7 never does when you o/c with any Tbred or Barton, you only get the 'Athlon at _ghz' which deflates your o/c'ing achievements. Well it does show the XP2500+ as XP3200+ if you switch from 11x166 to 11x200 but that's all. I thought perhaps it's a VIA vs nForce2 thing but then I heard that Asus A7N8X Deluxe and Epox nForce2 both apply the XP rating. I then figured it may be a brand thing (eg Giga, Asus and Epox do while Abit don't) but then I learned the Asus A7V8X-X KT400 didn't apply the XP rating when o/c'ed ... you just get the stock XP rating and the o/c'ed ghz quoted instead. So basically it seems there is no standard or pattern present, some mobos do and others don't. It would seem that the odd mobo of the odd manu choose not to apply an XP rating when you o/c. Just seems a shame really.

:cool: Anyway the XP rating is supposedly derived from some calculation based on the old Thunderbird Athlon but to anyone with basic knowledge of CPU perf it's abundently clear that AMD wished to avoid lawsuit type complications ... it's very apparent it is designed to give the Athlon a rating equivilent to P4. There are many ways to test this out, eg I rem the release of the XP2400+ where the XP rating was held back a notch, that just happened to coincide with the P4's move from 400FSB to 533FSB (and 512k too IIRC). It's been very hard for AMD to apply this rating and keep it in line. P4's give very different perf depending upon FSB (400, 533 & 800), memory used (PC133, Rambus, DDR, Dual Channel DDR) and amount of cache. They've done pretty well over all and it certainly saved their bacon, very few people would believe an Athlon at 1.67ghz 266FSB 256k cache could ever outperform a P4 2.0ghz 533FSB 512k L2 for example. With Dual Channel DDR, 800FSB and HT the P4 really does pull away from the AthlonXP but then I guess that's where Athlon64 steps in.

EDIT: Silly silly spealling mishtakes LOL!