First, there are advantages of smaller cylinders and shorter crank throws in terms of rotating mass - you can make a small light piston on a short stroke rev more easily than a big heavy piston on a long stroke.  
Then there's the size of the combustion chamber itself - in a big chamber, it takes longer for the wave front of combustion to spread from the spark plug and ignite the whole mixture, and the inlet charge has further to travel and more to fill, which has to take longer.

Smaller chambers also have disadvantages - the ratio of volume to wall surface and inlet tract diameter is smaller, so wall friction (both in terms of piston/cylinder drag and gas drag through the ports) has a proportionately greater effect on a small cylinder than on a large one.

For ultimate power, you want revs above all else so smaller is generally better.  For two engines of similar capacity aiming for driveable mid-range torque at low-ish revs, the optimum is debateable but you're certainly getting into lower efficiency when the cylinder size exceeds 500cc.  
On performance bikes, wanting power over 10,000rpm, specific outputs seem to be highest on high-revving motors with cylinders about 150-175cc.
For cars, aiming for more torque, and usable ranges up to 6000rpm, specific outputs seem to be best around 400-500cc cylinders.

It's not really fair to compare the Savage’s output with other bikes of similar capacity because it was marketed for a restricted horsepower market in Europe where some categories of new/learner riders are limited to a 32hp bike, so the Savage was built down to that power.  The carb, cam, exhaust, exhaust port and valves sizes are all restrictive

Without those built-in restrictions, a Savage should comfortably make 45-50hp.  

A 650 twin in the same state of tune would make more power and not much less torque.


1973 Triumph Bonneville 650 vertical twin 49hp at 7200rpm
1975 Yamaha XS650 vertical twin 51hp at 7500rpm  (37ftlb at 6000rpm)
1996 BMW F650 flat twin 50hp at 6800rpm      (44ftlb at 5500rpm)
1999 Suzuki SV650 V twin 69bhp,  45ft lb

And just to throw a cat among the pigeons:

1996 Suzuki Savage single 30hp at 5400rpm, (33 ftlb at 3400rpm)
1996 Virago 535 V-twin  44hp at 7500rpm  (35ftlb at 6000rpm)

Those 535 Virago's are really cool Bikes ... wish Yamahog would bring back a modern version of it (and don't say the 650 V-Star ... like the Bike, but it's too heavy)!

musicdorian wrote on 05/28/08 at 22:32:37:


For the love of god, don't let people put thoughts like that in your head.  

Instead of just blindly believing their clap-trap, go out and find out for your self.   A quick demo ride of both bikes would definitavely answer the question, because it would be you doing the answering.

If you ask here, we're likely to tell you that the Savage walks on water, gets 1,000,000 miles to the gallon, can do 1,000,000,000 miles per hour and all the talk of uncomfortable seats is hog wash.  But then again, we're kind of slanted...

Let me put it this way.  Last weekend, I rode 700 miles over 3 days.  Yes, I have better shocks on the rear and a better seat than stock, but at no time was my bike 'underpowered' or unable to keep up with traffic.  Quite the opposite.  Mrs. Skrapiron and I joined a 21 bike group ride from Loudonville, Ohio to Mount Vernon, Ohio, where we met my parents for dinner.  We rode mid pack and never once had any trouble keeping up with the Yamaha Raider that was leading the group.  That same weeked, I got nailed for 78 in a 65.  

The Savage can be described alot of different ways.  Underpowered isn't one of them......

Paladin. wrote on 05/29/08 at 03:56:26:

Right, this is about every single or twin 650cc bike I can find to compare.  I ddn’t include the Suzuki SV 650 because this is quite sporty and high revving and no really the same category of bike as we should compare with.

But, very clearly, the average torque of the twins is higher than that of the singles.

Year      Make      Model      Layout      Style      Hp      at rpm      ftlb      at rpm
                                               
1996      Suzuki      Savage 650      single      Cruiser      30      5400      33      3400
1999      Kawasaki      KLR650      single      Dirt/Trail      43      7000      35      5000
2001      Sachs      Roadser 650      single      Naked/retro      50      6700      41      6200
2000      Aprilia      Pegaso      single      Dirt/Trail      50            40      
                       650 single average      43.3hp      37.3ftlb
                                               
1973      Triumph      Bonneville 650      vertical twin      Classic      49      7200            
1975      Yamaha      XS650      vertical twin      Classic      51      7500      37      6000
1996      BMW      F650      flat twin      Tourer      50      6800      44      5500
1989      Honda      XRV650      parallel twin      Dirt/Trail      57      8000      44      6000
2001      Kawasaki      W650      parallel twin      Retro      50      7000      42      5600
2006      Yamaha      Dragstar 650      v twin      Cruiser      39      6500      37      5100
                       650 twin average      49.3hp      40.8ftlb
                                               
1999      Suzuki      SV650      v twin      Sports Tourer      69      9000      45      7400

If we exclude the Savage from the single averages (after all, it is deliberately restricted), the figures for singles rise to 48hp and 39ftlb average – close to the twins, but still below on torque as well as power..


If I were to add the following bikes’ details:

Year      Make      Model      Layout      Style      Hp      at rpm      ftlb      at rpm
1988      Honda      XR600      single      Dirt/Trail      45      6500      38      5500
1992      Yamaha      XT600      single      Dirt/Trail      45      6500      36      5500

And factor them up by 8.3% to equate their 600cc outputs to those at 650cc, and add them to the 650 single average, the average figures remain 48hp/39ftlb for the singles – still below the twins.

Maybe I’m missing something here, but I’m not quite seeing how singles give out more torque than multis.

If we look at 4 cylinder motors,

Year      Make      Model      Layout      Style      Hp      at rpm      ftlb      at rpm
1981      Honda      CB650 Custom      4 cylinder      Custom/Cruiser      63      9000      38      800
1977      Kawasaki      Z650      4 cylinder      Naked/sports      64      8500      42      7000
2005      Suzuki      GSF650      4 cylinder      Naked/sports      78      10000      44      7800
2007      Suzuki      GSX650      4 cylinder      Sports      86            46      
2005      Triumph      Daytona 650      5 cylinder      Sports      114      12500      50      11500
                       650 four average      81hp      44ftlb


No – I still don’t see it!!

bill67 wrote on 05/29/08 at 10:16:39:
Rpm is just a number/noise frequency.  Tractive force at the rear wheels is the source of acceleration.
That is indeed the result of torque - but it's torque factored by gearing, and if you produce that torque at higher rpm, you can use shorter gearing to produce harder acceleration.
You could have 2 engines producing exactly the same torque, but one at 3000rpm and one at 6000rpm.  The one at 6000rpm could run a gear ratio twice as advantageous, and give twice the acceleration - for the same torque.
That's why, contrary to common misunderstanding, it's actually power, not torque figure, which is important to racers (whether circuit or straight line).  Power and torque are actually just different ways of expressing the output of an engine – power is just torque multiplied by engine speed, so the higher up the rev range you produce your torque, the more power, and the more you can gear it to get force at the wheels.
It isn’t either power or torque that creates acceleration, it’s tractive force at the road/tyre interface, and that is the product of torque and gearing – and the gearing is dependent on where the engine develops its torque, ie its power.

  I don't like high rpm motors

KwakNut wrote on 05/29/08 at 16:43:17:


Hmmm, for a racer who isn't concerned about clutch replacement, gas mileage or drivability, I'd agree whole-heartedly, but.....

Since I don't fit that category, torque at a given RPM is significant.  You clutch locks up only when the engine and tranny are at the same speed, therefore...to take advantage of torque at a higher RPM you must have increased "slip" to take advantage of it either by tearing your clutch to shreds, or by use of a torque converter, since regardless of your gear ratio, the tranny is at 0 rpms and the engine is > 0 (if it's running.)  Either method is INEFFICIENT and UNHEALTHY for mechanical parts.  Again, no problem for racers.

There's a cost to everything, multiple cylinders increases complexity, weight, and size for a given displacement.  MPG will also drop.  

The gear ratio concept applies to racing, but not so much real life...
The Savage runs 4000 rpms at 60 mph stock.  What bike out there runs 16000 rpms at 60 mph in high gear?!?  Whatever gearing you choose, the engine that produces torque at the lowest RPMs will take advantage of it sooner and accelerate quicker (but have a lower top speed.)