I was asked this question on a forum yesterday. The 'throughput' here is the '[b]max.[/b] throughput'.
Below is my answer:
DSSS specified in the 802.11b standard provides a data rate of 5.5 and 11 Mbps, and only 3 non-overlapping channels (3 co-located APs) exist in the same space. This results in a max. throughput of 11 Mbps ? 3 APs = 33 Mbps.
FHSS specified in the original 802.11 standard provides a data rate of 1 and 2 Mbps. FHSS has a max. bandwidth of 79 MHz but you can't deploy 79 individual APs for sure. 10 to 12 co-located APs are recommended in the real world. This results in a max. throughput of 2 Mbps ? 12 APs = 24 Mbps.
Conclusion: DSSS has a better max. throughput than FHSS.
Ideally, with co-located FHSS systems each AP would have an orthogonal hopping sequence compared with the others. In other words, there should be no correlation at any time between any of the frequencies transmitted ( helping prevent mutual "co-channel" interference ). As the number of APs increase, there is the possibility of one AP "stepping on another". This could possibly happen if the speeds at which the TX synthesizers in the radios in each AP were able to change TX frequency had slight differences ( the actual frequency agility specs ).
I have seen a similar scenario in the past with direct sequence spread spectrum line of sight ( non Wi-Fi ) links in some of the countries where I have worked where unlicensed means just that !! Everybody and their aunt transmitting all over the place !. In theory, the orthogonality provided by the spreading codes should allow you to have a very large number of systems all transmitting at the same time and at the same frequency and on the same polarization without mutual interference. In reality, you could only get about half a dozen to a dozen stations doing that.
FHSS can be a tough old bird as far as some types of interference are concerned.
You sometimes see answers to test questions stating factors of "1/75"or "75 times" in relation to FHSS interference immunity. That is a gross simplication and quite simply misleading and wrong. Interference is one of the most complex issues in RF engineering. The coding and modulation schemes used, power levels used, frequency offsets etc etc are very important.
There are still a lot of military FHSS radio systems out there, mainly due to their ruggedness.